text
stringlengths 101
3.69k
| messages
listlengths 1
1
| outputs
dict |
|---|---|---|
In animals , amino acids are obtained through the consumption of foods containing protein . Ingested proteins are then broken down into amino acids through digestion , which typically involves denaturation of the protein through exposure to acid and hydrolysis by enzymes called proteases . Some ingested amino acids are used for protein biosynthesis , while others are converted to glucose through gluconeogenesis , or fed into the citric acid cycle . This use of protein as a fuel is particularly important under starvation conditions as it allows the body 's own proteins to be used to support life , particularly those found in muscle . Amino acids are also an important dietary source of nitrogen .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n In animals , amino acids are obtained through the consumption of foods containing protein . Ingested proteins are then broken down into amino acids through digestion , which typically involves denaturation of the protein through exposure to acid and hydrolysis by enzymes called proteases . Some ingested amino acids are used for protein biosynthesis , while others are converted to glucose through gluconeogenesis , or fed into the citric acid cycle . This use of protein as a fuel is particularly important under starvation conditions as it allows the body 's own proteins to be used to support life , particularly those found in muscle . Amino acids are also an important dietary source of nitrogen . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "In animals, <|start-PI|>amino acids are obtained through<|end-PI|> <|retrieval|> <|start-KFI|>the consumption of foods containing protein<|end-KFI|>. <|start-PI|>Ingested proteins are then broken down into amino acids through digestion, which typically involves<|end-PI|> <|retrieval|> <|start-KFI|>denaturation of the protein through exposure to acid and hydrolysis by enzymes called proteases<|end-KFI|>. <|start-PI|>Some ingested amino acids are used for protein biosynthesis, while others are<|end-PI|> <|retrieval|> <|start-KFI|>converted to glucose through gluconeogenesis, or fed into the citric acid cycle<|end-KFI|>. <|start-PI|>This use of protein as a fuel is particularly important under starvation conditions as it allows<|end-PI|> <|retrieval|> <|start-KFI|>the body's own proteins to be used to support life, particularly those found in muscle<|end-KFI|>. Amino acids are also an important dietary source of nitrogen.",
"none": null,
"results": [
{
"Explanation": "The PI sets up a query about how animals obtain amino acids, and the KFI provides the specific method. The <|retrieval|> token is inserted to indicate where additional information about amino acid sources might be needed.",
"Ideal Knowledge": "Information on animal nutrition, specifically detailing the sources of amino acids in an animal's diet, with emphasis on protein-rich foods.",
"Key Factual Information": "the consumption of foods containing protein",
"Prefix reformulated": "How do animals obtain amino acids?",
"Prefix-intent": "In animals, amino acids are obtained through",
"Verification": "The KFI directly answers how animals obtain amino acids, as prompted by the PI."
},
{
"Explanation": "The PI introduces the topic of protein digestion and sets up a query about the process, while the KFI provides specific details about the mechanisms involved. The <|retrieval|> token is inserted to indicate where additional information about protein digestion might be needed.",
"Ideal Knowledge": "Detailed explanation of the protein digestion process, including the role of stomach acid in denaturing proteins and the function of proteases in breaking down proteins into amino acids.",
"Key Factual Information": "denaturation of the protein through exposure to acid and hydrolysis by enzymes called proteases",
"Prefix reformulated": "What does the digestion of proteins typically involve?",
"Prefix-intent": "Ingested proteins are then broken down into amino acids through digestion, which typically involves",
"Verification": "The KFI directly describes the process of protein digestion, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the fate of amino acids not used for protein synthesis, and the KFI provides specific metabolic pathways. The <|retrieval|> token is inserted to indicate where additional information about amino acid metabolism might be needed.",
"Ideal Knowledge": "Information on amino acid metabolism, specifically detailing the processes of gluconeogenesis and the citric acid cycle, and how amino acids can be used as energy sources.",
"Key Factual Information": "converted to glucose through gluconeogenesis, or fed into the citric acid cycle",
"Prefix reformulated": "What happens to ingested amino acids that are not used for protein biosynthesis?",
"Prefix-intent": "Some ingested amino acids are used for protein biosynthesis, while others are",
"Verification": "The KFI directly answers what happens to amino acids not used for protein biosynthesis, as prompted by the PI."
},
{
"Explanation": "The PI introduces the importance of protein as fuel during starvation, and the KFI provides specific details about how the body uses its own proteins. The <|retrieval|> token is inserted to indicate where additional information about protein metabolism during starvation might be needed.",
"Ideal Knowledge": "Detailed information on the body's response to starvation, including the breakdown of muscle proteins for energy and the prioritization of essential bodily functions.",
"Key Factual Information": "the body's own proteins to be used to support life, particularly those found in muscle",
"Prefix reformulated": "How does the use of protein as fuel help during starvation?",
"Prefix-intent": "This use of protein as a fuel is particularly important under starvation conditions as it allows",
"Verification": "The KFI directly explains how protein is used as fuel during starvation, as prompted by the PI."
}
]
} |
Proteins were recognized as a distinct class of biological molecules in the eighteenth century by Antoine Fourcroy and others , distinguished by the molecules ' ability to coagulate or flocculate under treatments with heat or acid . Noted examples at the time included albumin from egg whites , blood serum albumin , fibrin , and wheat gluten .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Proteins were recognized as a distinct class of biological molecules in the eighteenth century by Antoine Fourcroy and others , distinguished by the molecules ' ability to coagulate or flocculate under treatments with heat or acid . Noted examples at the time included albumin from egg whites , blood serum albumin , fibrin , and wheat gluten . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Proteins were recognized as a distinct class of biological molecules in the eighteenth century by<|end-PI|> <|retrieval|> <|start-KFI|>Antoine Fourcroy and others<|end-KFI|> , <|start-PI|>distinguished by the molecules ' ability to<|end-PI|> <|retrieval|> <|start-KFI|>coagulate or flocculate under treatments with heat or acid<|end-KFI|> . <|start-PI|>Noted examples at the time included<|end-PI|> <|retrieval|> <|start-KFI|>albumin from egg whites , blood serum albumin , fibrin , and wheat gluten<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context for the discovery of proteins, and the KFI provides the key individuals involved. The <|retrieval|> token is inserted to indicate where additional information about the scientists might be needed.",
"Ideal Knowledge": "A detailed account of 18th-century scientists who contributed to the identification and classification of proteins, with a focus on Antoine Fourcroy's work and contributions.",
"Key Factual Information": "Antoine Fourcroy and others",
"Prefix reformulated": "Who recognized proteins as a distinct class of biological molecules in the eighteenth century?",
"Prefix-intent": "Proteins were recognized as a distinct class of biological molecules in the eighteenth century by",
"Verification": "The KFI directly provides the names of those who recognized proteins, as prompted by the PI."
},
{
"Explanation": "The PI introduces the distinguishing feature of proteins, and the KFI provides the specific molecular behavior. The <|retrieval|> token is inserted to indicate where additional information about protein behavior might be needed.",
"Ideal Knowledge": "Detailed explanation of protein denaturation processes, including coagulation and flocculation, and how these reactions occur under heat or acidic conditions.",
"Key Factual Information": "coagulate or flocculate under treatments with heat or acid",
"Prefix reformulated": "What ability distinguished proteins as a distinct class of biological molecules?",
"Prefix-intent": "Proteins were distinguished by the molecules ' ability to",
"Verification": "The KFI directly describes the distinguishing characteristic of proteins, as prompted by the PI."
},
{
"Explanation": "The PI sets up a list of protein examples, and the KFI provides specific proteins known at the time. The <|retrieval|> token is inserted to indicate where additional information about these protein examples might be needed.",
"Ideal Knowledge": "A comprehensive list of proteins identified in the 18th century, including their sources, properties, and the methods used to isolate and study them.",
"Key Factual Information": "albumin from egg whites , blood serum albumin , fibrin , and wheat gluten",
"Prefix reformulated": "What were some noted examples of proteins recognized in the eighteenth century?",
"Prefix-intent": "Noted examples at the time included",
"Verification": "The KFI directly lists examples of proteins, as prompted by the PI."
}
]
} |
Proteins were first described by the Dutch chemist Gerardus Johannes Mulder and named by the Swedish chemist Jöns Jacob Berzelius in 1838 . Mulder carried out elemental analysis of common proteins and found that nearly all proteins had the same empirical formula , C400H620N100O120P1S1 . He came to the erroneous conclusion that they might be composed of a single type of ( very large ) molecule . The term " protein " to describe these molecules was proposed by Mulder 's associate Berzelius ; protein is derived from the Greek word πρώτειος ( proteios ) , meaning " primary " , " in the lead " , or " standing in front " , + -in . Mulder went on to identify the products of protein degradation such as the amino acid leucine for which he found a ( nearly correct ) molecular weight of 131 Da .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Proteins were first described by the Dutch chemist Gerardus Johannes Mulder and named by the Swedish chemist Jöns Jacob Berzelius in 1838 . Mulder carried out elemental analysis of common proteins and found that nearly all proteins had the same empirical formula , C400H620N100O120P1S1 . He came to the erroneous conclusion that they might be composed of a single type of ( very large ) molecule . The term \" protein \" to describe these molecules was proposed by Mulder 's associate Berzelius ; protein is derived from the Greek word πρώτειος ( proteios ) , meaning \" primary \" , \" in the lead \" , or \" standing in front \" , + -in . Mulder went on to identify the products of protein degradation such as the amino acid leucine for which he found a ( nearly correct ) molecular weight of 131 Da . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Proteins were first described by the Dutch chemist<|end-PI|> <|retrieval|> <|start-KFI|>Gerardus Johannes Mulder<|end-KFI|> and <|start-PI|>Proteins were first named by the Swedish chemist<|end-PI|> <|retrieval|> <|start-KFI|>Jöns Jacob Berzelius in 1838<|end-KFI|>. <|start-PI|>Mulder carried out elemental analysis of common proteins and found that nearly all proteins had the same empirical formula,<|end-PI|> <|retrieval|> <|start-KFI|>C400H620N100O120P1S1<|end-KFI|>. He came to the erroneous conclusion that they might be composed of a single type of (very large) molecule. <|start-PI|>The term \"protein\" to describe these molecules was proposed by Mulder's associate Berzelius; protein is derived from the Greek word<|end-PI|> <|retrieval|> <|start-KFI|>πρώτειος (proteios), meaning \"primary\", \"in the lead\", or \"standing in front\"<|end-KFI|>, + -in. <|start-PI|>Mulder went on to identify the products of protein degradation such as the amino acid leucine for which he found a (nearly correct) molecular weight of<|end-PI|> <|retrieval|> <|start-KFI|>131 Da<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context for the discovery of proteins, and the KFI provides the specific name of the scientist. The <|retrieval|> token is inserted to indicate where additional information about the scientist might be needed.",
"Ideal Knowledge": "A brief biography of Gerardus Johannes Mulder, highlighting his contributions to protein chemistry and the year he first described proteins.",
"Key Factual Information": "Gerardus Johannes Mulder",
"Prefix reformulated": "Who first described proteins?",
"Prefix-intent": "Proteins were first described by the Dutch chemist",
"Verification": "The KFI directly provides the name of the chemist who first described proteins, as prompted by the PI."
},
{
"Explanation": "The PI introduces the naming of proteins, and the KFI provides the specific name of the scientist and the year. The <|retrieval|> token is inserted to indicate where additional information about Berzelius and the naming event might be needed.",
"Ideal Knowledge": "A brief biography of Jöns Jacob Berzelius, focusing on his role in naming proteins and his collaboration with Mulder in 1838.",
"Key Factual Information": "Jöns Jacob Berzelius in 1838",
"Prefix reformulated": "Who first named proteins?",
"Prefix-intent": "Proteins were first named by the Swedish chemist",
"Verification": "The KFI directly provides the name of the chemist who named proteins and the year, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of Mulder's analysis, and the KFI provides the specific empirical formula he discovered. The <|retrieval|> token is inserted to indicate where additional information about the analysis and formula might be needed.",
"Ideal Knowledge": "Detailed information about Mulder's elemental analysis of proteins, including his methodology and the significance of finding a common empirical formula for most proteins.",
"Key Factual Information": "C400H620N100O120P1S1",
"Prefix reformulated": "What empirical formula did Mulder find for nearly all proteins?",
"Prefix-intent": "Mulder carried out elemental analysis of common proteins and found that nearly all proteins had the same empirical formula,",
"Verification": "The KFI directly provides the empirical formula that Mulder found for proteins, as prompted by the PI."
},
{
"Explanation": "The PI introduces the origin of the term 'protein', and the KFI provides the specific Greek word and its meanings. The <|retrieval|> token is inserted to indicate where additional etymological information might be needed.",
"Ideal Knowledge": "Etymological information about the term 'protein', including its Greek root, pronunciation, and the reasoning behind Berzelius' choice of this term.",
"Key Factual Information": "πρώτειος (proteios), meaning \"primary\", \"in the lead\", or \"standing in front\"",
"Prefix reformulated": "What is the Greek word from which the term 'protein' is derived?",
"Prefix-intent": "The term \"protein\" to describe these molecules was proposed by Mulder's associate Berzelius; protein is derived from the Greek word",
"Verification": "The KFI directly provides the Greek word and its meanings from which 'protein' is derived, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of Mulder's work on protein degradation and leucine, and the KFI provides the specific molecular weight he determined. The <|retrieval|> token is inserted to indicate where additional information about leucine and molecular weight determination might be needed.",
"Ideal Knowledge": "Detailed information about Mulder's work on protein degradation, focusing on his discovery of leucine and the accuracy of his molecular weight determination compared to modern measurements.",
"Key Factual Information": "131 Da",
"Prefix reformulated": "What molecular weight did Mulder find for the amino acid leucine?",
"Prefix-intent": "Mulder went on to identify the products of protein degradation such as the amino acid leucine for which he found a (nearly correct) molecular weight of",
"Verification": "The KFI directly provides the molecular weight that Mulder found for leucine, as prompted by the PI."
}
]
} |
Early nutritional scientists such as the German Carl von Voit believed that protein was the most important nutrient for maintaining the structure of the body , because it was generally believed that " flesh makes flesh . " Karl Heinrich Ritthausen extended known protein forms with the identification of glutamic acid . At the Connecticut Agricultural Experiment Station a detailed review of the vegetable proteins was compiled by Thomas Burr Osborne . Working with Lafayette Mendel and applying Liebig 's law of the minimum in feeding laboratory rats , the nutritionally essential amino acids were established . The work was continued and communicated by William Cumming Rose . The understanding of proteins as polypeptides came through the work of Franz Hofmeister and Hermann Emil Fischer . The central role of proteins as enzymes in living organisms was not fully appreciated until 1926 , when James B. Sumner showed that the enzyme urease was in fact a protein .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Early nutritional scientists such as the German Carl von Voit believed that protein was the most important nutrient for maintaining the structure of the body , because it was generally believed that \" flesh makes flesh . \" Karl Heinrich Ritthausen extended known protein forms with the identification of glutamic acid . At the Connecticut Agricultural Experiment Station a detailed review of the vegetable proteins was compiled by Thomas Burr Osborne . Working with Lafayette Mendel and applying Liebig 's law of the minimum in feeding laboratory rats , the nutritionally essential amino acids were established . The work was continued and communicated by William Cumming Rose . The understanding of proteins as polypeptides came through the work of Franz Hofmeister and Hermann Emil Fischer . The central role of proteins as enzymes in living organisms was not fully appreciated until 1926 , when James B. Sumner showed that the enzyme urease was in fact a protein . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "Early nutritional scientists such as the German Carl von Voit believed that protein was the most important nutrient for maintaining the structure of the body , because <|start-PI|> Early nutritional scientists such as the German Carl von Voit believed that protein was the most important nutrient for maintaining the structure of the body , because <|end-PI|> <|retrieval|> <|start-KFI|> it was generally believed that \" flesh makes flesh .\" <|end-KFI|> <|start-PI|> Karl Heinrich Ritthausen extended known protein forms with the identification of <|end-PI|> <|retrieval|> <|start-KFI|> glutamic acid <|end-KFI|> . <|start-PI|> At the Connecticut Agricultural Experiment Station a detailed review of the vegetable proteins was compiled by <|end-PI|> <|retrieval|> <|start-KFI|> Thomas Burr Osborne <|end-KFI|> . <|start-PI|> Working with Lafayette Mendel and applying Liebig 's law of the minimum in feeding laboratory rats , <|end-PI|> <|retrieval|> <|start-KFI|> the nutritionally essential amino acids were established <|end-KFI|> . The work was continued and communicated by William Cumming Rose . <|start-PI|> The understanding of proteins as polypeptides came through the work of <|end-PI|> <|retrieval|> <|start-KFI|> Franz Hofmeister and Hermann Emil Fischer <|end-KFI|> . <|start-PI|> The central role of proteins as enzymes in living organisms was not fully appreciated until 1926 , when <|end-PI|> <|retrieval|> <|start-KFI|> James B. Sumner showed that the enzyme urease was in fact a protein <|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the belief of early nutritional scientists, while the KFI provides the specific reasoning. The <|retrieval|> token is inserted to indicate where additional historical context might be needed.",
"Ideal Knowledge": "Historical context of early nutritional science, focusing on the beliefs about protein's role in body structure and the origins of the 'flesh makes flesh' concept.",
"Key Factual Information": "it was generally believed that \" flesh makes flesh .\"",
"Prefix reformulated": "Why did early nutritional scientists like Carl von Voit believe protein was the most important nutrient for maintaining body structure?",
"Prefix-intent": "Early nutritional scientists such as the German Carl von Voit believed that protein was the most important nutrient for maintaining the structure of the body , because",
"Verification": "The KFI directly provides the reasoning behind the belief stated in the PI."
},
{
"Explanation": "The PI sets up Ritthausen's contribution, and the KFI provides the specific discovery. The <|retrieval|> token is inserted to indicate where additional information about this discovery might be needed.",
"Ideal Knowledge": "Detailed information about Karl Heinrich Ritthausen's work on protein forms, specifically his discovery and identification of glutamic acid and its significance in protein research.",
"Key Factual Information": "glutamic acid",
"Prefix reformulated": "What protein form did Karl Heinrich Ritthausen identify to extend known protein forms?",
"Prefix-intent": "Karl Heinrich Ritthausen extended known protein forms with the identification of",
"Verification": "The KFI directly provides the specific protein form identified by Ritthausen, as prompted by the PI."
},
{
"Explanation": "The PI introduces the review and its location, while the KFI provides the name of the compiler. The <|retrieval|> token is inserted to indicate where additional information about Osborne's work might be needed.",
"Ideal Knowledge": "Information about Thomas Burr Osborne's work at the Connecticut Agricultural Experiment Station, focusing on his comprehensive review of vegetable proteins and its impact on protein research.",
"Key Factual Information": "Thomas Burr Osborne",
"Prefix reformulated": "Who compiled a detailed review of vegetable proteins at the Connecticut Agricultural Experiment Station?",
"Prefix-intent": "At the Connecticut Agricultural Experiment Station a detailed review of the vegetable proteins was compiled by",
"Verification": "The KFI directly provides the name of the person who compiled the review, as prompted by the PI."
},
{
"Explanation": "The PI describes the research approach, and the KFI provides the significant result. The <|retrieval|> token is inserted to indicate where additional information about this research might be needed.",
"Ideal Knowledge": "Detailed information about the collaborative work of Lafayette Mendel and others, explaining how they applied Liebig's law of the minimum in rat feeding experiments to determine nutritionally essential amino acids.",
"Key Factual Information": "the nutritionally essential amino acids were established",
"Prefix reformulated": "What was established by working with Lafayette Mendel and applying Liebig's law of the minimum in feeding laboratory rats?",
"Prefix-intent": "Working with Lafayette Mendel and applying Liebig 's law of the minimum in feeding laboratory rats ,",
"Verification": "The KFI directly provides the outcome of the work described in the PI."
},
{
"Explanation": "The PI introduces the concept of proteins as polypeptides, and the KFI provides the names of the key contributors. The <|retrieval|> token is inserted to indicate where additional information about their work might be needed.",
"Ideal Knowledge": "Biographical information and research contributions of Franz Hofmeister and Hermann Emil Fischer, specifically their work on understanding proteins as polypeptides and its impact on protein science.",
"Key Factual Information": "Franz Hofmeister and Hermann Emil Fischer",
"Prefix reformulated": "Who contributed to the understanding of proteins as polypeptides?",
"Prefix-intent": "The understanding of proteins as polypeptides came through the work of",
"Verification": "The KFI directly provides the names of the scientists who contributed to this understanding, as prompted by the PI."
},
{
"Explanation": "The PI sets up the historical context, and the KFI provides the specific discovery that led to a new understanding. The <|retrieval|> token is inserted to indicate where additional information about this breakthrough might be needed.",
"Ideal Knowledge": "Detailed information about James B. Sumner's 1926 discovery, including his methods for demonstrating that urease was a protein and how this finding changed the understanding of proteins' role in living organisms.",
"Key Factual Information": "James B. Sumner showed that the enzyme urease was in fact a protein",
"Prefix reformulated": "What happened in 1926 that led to the appreciation of proteins' central role as enzymes in living organisms?",
"Prefix-intent": "The central role of proteins as enzymes in living organisms was not fully appreciated until 1926 , when",
"Verification": "The KFI directly provides the specific discovery and the scientist responsible, as prompted by the PI."
}
]
} |
The difficulty in purifying proteins in large quantities made them very difficult for early protein biochemists to study . Hence , early studies focused on proteins that could be purified in large quantities , e.g. , those of blood , egg white , various toxins , and digestive / metabolic enzymes obtained from slaughterhouses . In the 1950s , the Armour Hot Dog Co. purified 1 kg of pure bovine pancreatic ribonuclease A and made it freely available to scientists ; this gesture helped ribonuclease A become a major target for biochemical study for the following decades .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n The difficulty in purifying proteins in large quantities made them very difficult for early protein biochemists to study . Hence , early studies focused on proteins that could be purified in large quantities , e.g. , those of blood , egg white , various toxins , and digestive / metabolic enzymes obtained from slaughterhouses . In the 1950s , the Armour Hot Dog Co. purified 1 kg of pure bovine pancreatic ribonuclease A and made it freely available to scientists ; this gesture helped ribonuclease A become a major target for biochemical study for the following decades . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "The difficulty in purifying proteins in large quantities made them <|start-PI|> very difficult for early protein biochemists to study <|end-PI|> <|retrieval|> <|start-KFI|> very difficult for early protein biochemists to study <|end-KFI|> . Hence , early studies focused on proteins that could be purified in large quantities , e.g. , <|start-PI|> those of blood , egg white , various toxins , and digestive / metabolic enzymes obtained from slaughterhouses <|end-PI|> <|retrieval|> <|start-KFI|> those of blood , egg white , various toxins , and digestive / metabolic enzymes obtained from slaughterhouses <|end-KFI|> . In the 1950s , the Armour Hot Dog Co. purified 1 kg of pure bovine pancreatic ribonuclease A and <|start-PI|> made it freely available to scientists <|end-PI|> <|retrieval|> <|start-KFI|> made it freely available to scientists <|end-KFI|> ; this gesture helped ribonuclease A become <|start-PI|> a major target for biochemical study for the following decades <|end-PI|> <|retrieval|> <|start-KFI|> a major target for biochemical study for the following decades <|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the challenge of protein purification, while the KFI explains its direct impact on early biochemists. The <|retrieval|> token is inserted to indicate where additional information about these difficulties might be needed.",
"Ideal Knowledge": "Information on the challenges faced by early protein biochemists, specifically detailing the limitations imposed by protein purification difficulties and their impact on research methodologies.",
"Key Factual Information": "very difficult for early protein biochemists to study",
"Prefix reformulated": "How did the difficulty in purifying proteins in large quantities affect early protein biochemists?",
"Prefix-intent": "The difficulty in purifying proteins in large quantities made them",
"Verification": "The KFI directly addresses the impact of protein purification difficulties on early biochemists' studies, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context for early protein studies, and the KFI provides specific examples of proteins that were studied. The <|retrieval|> token is inserted to indicate where additional information about these protein sources might be needed.",
"Ideal Knowledge": "A comprehensive list of protein sources that were readily available in large quantities for early biochemical studies, including details on why these specific proteins were easier to purify and study.",
"Key Factual Information": "those of blood , egg white , various toxins , and digestive / metabolic enzymes obtained from slaughterhouses",
"Prefix reformulated": "What types of proteins did early studies focus on due to purification challenges?",
"Prefix-intent": "Hence , early studies focused on proteins that could be purified in large quantities , e.g. ,",
"Verification": "The KFI directly lists the types of proteins that were studied due to their availability in large quantities, as prompted by the PI."
},
{
"Explanation": "The PI introduces the action taken by the Armour Hot Dog Co., and the KFI explains what they did with the purified protein. The <|retrieval|> token is inserted to indicate where additional information about this event and its significance might be needed.",
"Ideal Knowledge": "Detailed information about the Armour Hot Dog Co.'s contribution to biochemical research in the 1950s, including the process of purifying ribonuclease A, the significance of making it freely available, and its impact on the scientific community.",
"Key Factual Information": "made it freely available to scientists",
"Prefix reformulated": "What did the Armour Hot Dog Co. do with the purified bovine pancreatic ribonuclease A in the 1950s?",
"Prefix-intent": "In the 1950s , the Armour Hot Dog Co. purified 1 kg of pure bovine pancreatic ribonuclease A and",
"Verification": "The KFI directly states what the company did with the purified protein, as prompted by the PI."
},
{
"Explanation": "The PI sets up the impact of the company's action, and the KFI explains the long-term consequence for biochemical studies. The <|retrieval|> token is inserted to indicate where additional information about the significance of ribonuclease A in biochemical research might be needed.",
"Ideal Knowledge": "Information on the long-term effects of the Armour Hot Dog Co.'s donation of ribonuclease A, including specific studies conducted, breakthroughs achieved, and how it shaped the field of biochemistry in the following decades.",
"Key Factual Information": "a major target for biochemical study for the following decades",
"Prefix reformulated": "How did the Armour Hot Dog Co.'s gesture affect the study of ribonuclease A?",
"Prefix-intent": "this gesture helped ribonuclease A become",
"Verification": "The KFI directly states the impact of the company's action on the study of ribonuclease A, as prompted by the PI."
}
]
} |
Linus Pauling is credited with the successful prediction of regular protein secondary structures based on hydrogen bonding , an idea first put forth by William Astbury in 1933 . Later work by Walter Kauzmann on denaturation , based partly on previous studies by Kaj Linderstrøm @-@ Lang , contributed an understanding of protein folding and structure mediated by hydrophobic interactions .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Linus Pauling is credited with the successful prediction of regular protein secondary structures based on hydrogen bonding , an idea first put forth by William Astbury in 1933 . Later work by Walter Kauzmann on denaturation , based partly on previous studies by Kaj Linderstrøm @-@ Lang , contributed an understanding of protein folding and structure mediated by hydrophobic interactions . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
The first protein to be sequenced was insulin , by Frederick Sanger , in 1949 . Sanger correctly determined the amino acid sequence of insulin , thus conclusively demonstrating that proteins consisted of linear polymers of amino acids rather than branched chains , colloids , or cyclols . He won the Nobel Prize for this achievement in 1958 .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n The first protein to be sequenced was insulin , by Frederick Sanger , in 1949 . Sanger correctly determined the amino acid sequence of insulin , thus conclusively demonstrating that proteins consisted of linear polymers of amino acids rather than branched chains , colloids , or cyclols . He won the Nobel Prize for this achievement in 1958 . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
The first protein structures to be solved were hemoglobin and myoglobin , by Max Perutz and Sir John Cowdery Kendrew , respectively , in 1958 . As of 2016 , the Protein Data Bank has over 115 @,@ 000 atomic @-@ resolution structures of proteins . In more recent times , cryo @-@ electron microscopy of large macromolecular assemblies and computational protein structure prediction of small protein domains are two methods approaching atomic resolution .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n The first protein structures to be solved were hemoglobin and myoglobin , by Max Perutz and Sir John Cowdery Kendrew , respectively , in 1958 . As of 2016 , the Protein Data Bank has over 115 @,@ 000 atomic @-@ resolution structures of proteins . In more recent times , cryo @-@ electron microscopy of large macromolecular assemblies and computational protein structure prediction of small protein domains are two methods approaching atomic resolution . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
Comparative Toxicogenomics Database curates protein – chemical interactions , as well as gene / protein – disease relationships and chemical @-@ disease relationships .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Comparative Toxicogenomics Database curates protein – chemical interactions , as well as gene / protein – disease relationships and chemical @-@ disease relationships . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Comparative Toxicogenomics Database curates<|end-PI|> <|retrieval|> <|start-KFI|>protein – chemical interactions , as well as gene / protein – disease relationships and chemical @-@ disease relationships<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the Comparative Toxicogenomics Database and sets up a query about what it curates, while the KFI provides the specific types of information it contains. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "A detailed list of the types of data curated by the Comparative Toxicogenomics Database, including specific examples of protein-chemical interactions, gene/protein-disease relationships, and chemical-disease relationships.",
"Key Factual Information": "protein – chemical interactions , as well as gene / protein – disease relationships and chemical @-@ disease relationships",
"Prefix reformulated": "What does the Comparative Toxicogenomics Database curate?",
"Prefix-intent": "Comparative Toxicogenomics Database curates",
"Verification": "The KFI directly provides the types of information curated by the database, as prompted by the PI."
}
]
} |
Protein Databank in Europe ( see also PDBeQuips , short articles and tutorials on interesting PDB structures )
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Protein Databank in Europe ( see also PDBeQuips , short articles and tutorials on interesting PDB structures ) \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Protein Databank in Europe ( see also PDBeQuips ,<|end-PI|> <|retrieval|> <|start-KFI|>short articles and tutorials on interesting PDB structures<|end-KFI|>)",
"none": null,
"results": [
{
"Explanation": "The PI introduces the Protein Databank in Europe and mentions PDBeQuips, setting up a query about what PDBeQuips is. The KFI provides a brief description of PDBeQuips' content. The <|retrieval|> token is inserted to indicate where additional information about PDBeQuips might be needed to support the description.",
"Ideal Knowledge": "Detailed information about PDBeQuips, including its purpose, content types, and relationship to the Protein Databank in Europe. This should include examples of the short articles and tutorials it offers, as well as how it relates to PDB structures.",
"Key Factual Information": "short articles and tutorials on interesting PDB structures",
"Prefix reformulated": "What is PDBeQuips in relation to the Protein Databank in Europe?",
"Prefix-intent": "Protein Databank in Europe ( see also PDBeQuips ,",
"Verification": "The KFI directly describes what PDBeQuips is, as prompted by the PI."
}
]
} |
Research Collaboratory for Structural Bioinformatics ( see also Molecule of the Month , presenting short accounts on selected proteins from the PDB )
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Research Collaboratory for Structural Bioinformatics ( see also Molecule of the Month , presenting short accounts on selected proteins from the PDB ) \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Research Collaboratory for Structural Bioinformatics<|end-PI|> <|retrieval|> <|start-KFI|>( see also Molecule of the Month , presenting short accounts on selected proteins from the PDB )<|end-KFI|>",
"none": null,
"results": [
{
"Explanation": "The PI introduces the Research Collaboratory for Structural Bioinformatics, and the KFI provides additional context about its associated features. The <|retrieval|> token is inserted to indicate where additional information about the organization might be needed.",
"Ideal Knowledge": "Detailed information about the Research Collaboratory for Structural Bioinformatics, including its purpose, activities, and associated resources like the Molecule of the Month feature and the Protein Data Bank (PDB).",
"Key Factual Information": "( see also Molecule of the Month , presenting short accounts on selected proteins from the PDB )",
"Prefix reformulated": "What is the Research Collaboratory for Structural Bioinformatics?",
"Prefix-intent": "Research Collaboratory for Structural Bioinformatics",
"Verification": "The KFI provides additional information about the Research Collaboratory for Structural Bioinformatics, as prompted by the PI."
}
]
} |
Proteopedia – Life in 3D : rotatable , zoomable 3D model with wiki annotations for every known protein molecular structure .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Proteopedia – Life in 3D : rotatable , zoomable 3D model with wiki annotations for every known protein molecular structure . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Proteopedia – Life in 3D : rotatable , zoomable 3D model with wiki annotations for<|end-PI|> <|retrieval|> <|start-KFI|>every known protein molecular structure<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces Proteopedia and its features, setting up a query about what it covers. The KFI provides the specific information about its comprehensive coverage of protein structures. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement about the extent of Proteopedia's database.",
"Ideal Knowledge": "A comprehensive description of Proteopedia, including its features such as rotatable and zoomable 3D models, wiki annotations, and its coverage of all known protein molecular structures. This should also include information about the database's scope and how it integrates structural data with annotations.",
"Key Factual Information": "every known protein molecular structure",
"Prefix reformulated": "What does Proteopedia – Life in 3D provide 3D models and wiki annotations for?",
"Prefix-intent": "Proteopedia – Life in 3D : rotatable , zoomable 3D model with wiki annotations for",
"Verification": "The KFI directly provides information about what Proteopedia offers 3D models and annotations for, as prompted by the PI."
}
]
} |
neXtProt – Exploring the universe of human proteins : human @-@ centric protein knowledge resource
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n neXtProt – Exploring the universe of human proteins : human @-@ centric protein knowledge resource \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>neXtProt – Exploring the universe of<|end-PI|> <|retrieval|> <|start-KFI|>human proteins<|end-KFI|> : <|start-PI|>neXtProt – Exploring the universe of human proteins :<|end-PI|> <|retrieval|> <|start-KFI|>human @-@ centric protein knowledge resource<|end-KFI|>",
"none": null,
"results": [
{
"Explanation": "The PI introduces neXtProt and sets up a query about what it explores, while the KFI provides the specific focus. The <|retrieval|> token is inserted to indicate where additional information about neXtProt's purpose might be needed.",
"Ideal Knowledge": "A detailed description of neXtProt, explaining that it is a comprehensive database or platform dedicated to exploring and cataloging human proteins, their functions, and interactions.",
"Key Factual Information": "human proteins",
"Prefix reformulated": "What is neXtProt exploring?",
"Prefix-intent": "neXtProt – Exploring the universe of",
"Verification": "The KFI directly provides the subject of exploration mentioned in the PI."
},
{
"Explanation": "The PI sets up the context of neXtProt and human proteins, while the KFI provides a concise description of what neXtProt actually is. The <|retrieval|> token is inserted to indicate where additional information about neXtProt's nature and purpose might be needed.",
"Ideal Knowledge": "An explanation of neXtProt's role as a comprehensive, human-focused protein database, detailing its features, data sources, and how it serves as a centralized resource for protein-related information in human biology.",
"Key Factual Information": "human @-@ centric protein knowledge resource",
"Prefix reformulated": "What is neXtProt in relation to human proteins?",
"Prefix-intent": "neXtProt – Exploring the universe of human proteins :",
"Verification": "The KFI directly describes what neXtProt is in relation to human proteins, as prompted by the PI."
}
]
} |
Multi @-@ Omics Profiling Expression Database : MOPED human and model organism protein / gene knowledge and expression data
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Multi @-@ Omics Profiling Expression Database : MOPED human and model organism protein / gene knowledge and expression data \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Multi @-@ Omics Profiling Expression Database : MOPED<|end-PI|> <|retrieval|> <|start-KFI|>human and model organism protein / gene knowledge and expression data<|end-KFI|>",
"none": null,
"results": [
{
"Explanation": "The PI introduces MOPED and sets up a query about its nature, while the KFI provides specific information about the database's contents. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "Detailed information about MOPED, including its full name, purpose, and the types of data it contains, specifically mentioning its focus on human and model organism protein and gene data.",
"Key Factual Information": "human and model organism protein / gene knowledge and expression data",
"Prefix reformulated": "What is MOPED (Multi-Omics Profiling Expression Database)?",
"Prefix-intent": "Multi @-@ Omics Profiling Expression Database : MOPED",
"Verification": "The KFI directly describes the content of the MOPED database, as prompted by the PI."
}
]
} |
" An Introduction to Proteins " from HOPES ( Huntington 's Disease Outreach Project for Education at Stanford )
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n \" An Introduction to Proteins \" from HOPES ( Huntington 's Disease Outreach Project for Education at Stanford ) \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|> An Introduction to Proteins <|end-PI|> <|retrieval|> <|start-KFI|> An Introduction to Proteins <|end-KFI|> <|start-PI|> from HOPES <|end-PI|> <|retrieval|> <|start-KFI|> ( Huntington 's Disease Outreach Project for Education at Stanford ) <|end-KFI|>",
"none": null,
"results": [
{
"Explanation": "The PI serves as the title of the article, which is also the KFI. The <|retrieval|> token is inserted after the title to indicate where additional information about the content might be needed.",
"Ideal Knowledge": "A list of educational articles or resources about proteins, specifically mentioning 'An Introduction to Proteins' as a title of an article or section.",
"Key Factual Information": "An Introduction to Proteins",
"Prefix reformulated": "What is the title of this article?",
"Prefix-intent": "An Introduction to Proteins",
"Verification": "The KFI directly provides the title information requested in the PI."
},
{
"Explanation": "The PI introduces the source of the article using an acronym, while the KFI provides the full name of the organization. The <|retrieval|> token is inserted to indicate where additional information about HOPES might be needed.",
"Ideal Knowledge": "Information about HOPES, including its full name, purpose, and affiliation with Stanford University, specifically in relation to Huntington's Disease education and outreach.",
"Key Factual Information": "Huntington's Disease Outreach Project for Education at Stanford",
"Prefix reformulated": "What is the source of this article?",
"Prefix-intent": "from HOPES",
"Verification": "The KFI directly provides the full name of the organization mentioned in the PI."
}
]
} |
Risa Oribe ( 織部 里沙 , Oribe Risa , born June 24 , 1987 ) , better known by her stage name LiSA ( an acronym of Love is Same All ) , is a Japanese pop singer @-@ songwriter from Seki , Gifu , signed to Aniplex under Sony Music Artists . After aspiring to become a musician early in life , she started her musical career as the vocalist of the indie band Chucky . Following Chucky 's disbandment in 2005 , LiSA moved to Tokyo in order to pursue a solo career , making her major debut in 2010 singing songs for the anime television series Angel Beats ! as one of two vocalists for the fictional band Girls Dead Monster . In April 2011 , she made her solo debut with the release of her mini @-@ album Letters to U. She performed at Animelo Summer Live in August 2010 , Anime Expo in 2012 , and is a regular guest at Anime Festival Asia .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Risa Oribe ( 織部 里沙 , Oribe Risa , born June 24 , 1987 ) , better known by her stage name LiSA ( an acronym of Love is Same All ) , is a Japanese pop singer @-@ songwriter from Seki , Gifu , signed to Aniplex under Sony Music Artists . After aspiring to become a musician early in life , she started her musical career as the vocalist of the indie band Chucky . Following Chucky 's disbandment in 2005 , LiSA moved to Tokyo in order to pursue a solo career , making her major debut in 2010 singing songs for the anime television series Angel Beats ! as one of two vocalists for the fictional band Girls Dead Monster . In April 2011 , she made her solo debut with the release of her mini @-@ album Letters to U. She performed at Animelo Summer Live in August 2010 , Anime Expo in 2012 , and is a regular guest at Anime Festival Asia . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Risa Oribe ( 織部 里沙 , Oribe Risa , born<|end-PI|> <|retrieval|> <|start-KFI|>June 24 , 1987<|end-KFI|> ) , <|start-PI|>better known by her stage name LiSA ( an acronym of<|end-PI|> <|retrieval|> <|start-KFI|>Love is Same All<|end-KFI|> ) , <|start-PI|>LiSA ( an acronym of Love is Same All ) , is a Japanese pop singer @-@ songwriter from<|end-PI|> <|retrieval|> <|start-KFI|>Seki , Gifu<|end-KFI|> , signed to Aniplex under Sony Music Artists . After aspiring to become a musician early in life , she started her musical career as the vocalist of the indie band Chucky . <|start-PI|>Following Chucky 's disbandment in 2005 , LiSA moved to<|end-PI|> <|retrieval|> <|start-KFI|>Tokyo in order to pursue a solo career<|end-KFI|> , <|start-PI|>LiSA moved to Tokyo in order to pursue a solo career , making her major debut in<|end-PI|> <|retrieval|> <|start-KFI|>2010 singing songs for the anime television series Angel Beats ! as one of two vocalists for the fictional band Girls Dead Monster<|end-KFI|> . <|start-PI|>In April 2011 , she made her solo debut with the release of her<|end-PI|> <|retrieval|> <|start-KFI|>mini @-@ album Letters to U<|end-KFI|> . She performed at Animelo Summer Live in August 2010 , Anime Expo in 2012 , and is a regular guest at Anime Festival Asia .",
"none": null,
"results": [
{
"Explanation": "The PI introduces Risa Oribe and sets up a query about her birth date, while the KFI provides the specific date. The <|retrieval|> token is inserted to indicate where additional biographical information might be needed.",
"Ideal Knowledge": "A detailed biography of Risa Oribe, including her full name in both Japanese and English, and her exact date of birth.",
"Key Factual Information": "June 24 , 1987",
"Prefix reformulated": "When was Risa Oribe born?",
"Prefix-intent": "Risa Oribe ( 織部 里沙 , Oribe Risa , born",
"Verification": "The KFI directly provides the birth date requested in the PI."
},
{
"Explanation": "The PI introduces LiSA's stage name and sets up a query about its meaning, while the KFI provides the full phrase behind the acronym. The <|retrieval|> token is inserted to indicate where additional information about her stage name might be needed.",
"Ideal Knowledge": "Information about LiSA's stage name, including its origin, meaning, and any background on why she chose this name.",
"Key Factual Information": "Love is Same All",
"Prefix reformulated": "What does LiSA's stage name stand for?",
"Prefix-intent": "better known by her stage name LiSA ( an acronym of",
"Verification": "The KFI directly provides the meaning of the acronym LiSA, as prompted by the PI."
},
{
"Explanation": "The PI introduces LiSA's profession and sets up a query about her origin, while the KFI provides the specific location. The <|retrieval|> token is inserted to indicate where additional information about Seki, Gifu might be needed.",
"Ideal Knowledge": "Detailed information about Seki, Gifu, including its location in Japan, population, and any notable features or cultural significance, especially in relation to the music industry.",
"Key Factual Information": "Seki , Gifu",
"Prefix reformulated": "Where is LiSA from?",
"Prefix-intent": "LiSA ( an acronym of Love is Same All ) , is a Japanese pop singer @-@ songwriter from",
"Verification": "The KFI directly provides LiSA's place of origin, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about LiSA's actions after her band disbanded, and the KFI provides information about her move and career goals. The <|retrieval|> token is inserted to indicate where additional information about her career transition might be needed.",
"Ideal Knowledge": "Information about LiSA's career progression, including details about her time with Chucky, the band's disbandment, and her subsequent move to Tokyo to pursue a solo career.",
"Key Factual Information": "Tokyo in order to pursue a solo career",
"Prefix reformulated": "Where did LiSA move after Chucky's disbandment in 2005?",
"Prefix-intent": "Following Chucky 's disbandment in 2005 , LiSA moved to",
"Verification": "The KFI directly provides information about LiSA's move and its purpose, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about LiSA's major debut, and the KFI provides specific details about when and how it occurred. The <|retrieval|> token is inserted to indicate where additional information about her debut project might be needed.",
"Ideal Knowledge": "Detailed information about LiSA's major debut, including the year 2010, her role in the anime series Angel Beats!, and her participation in the fictional band Girls Dead Monster.",
"Key Factual Information": "2010 singing songs for the anime television series Angel Beats ! as one of two vocalists for the fictional band Girls Dead Monster",
"Prefix reformulated": "When did LiSA make her major debut?",
"Prefix-intent": "LiSA moved to Tokyo in order to pursue a solo career , making her major debut in",
"Verification": "The KFI directly provides information about LiSA's major debut, including the year and specific project, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about LiSA's solo debut release, and the KFI provides the specific details of the mini-album. The <|retrieval|> token is inserted to indicate where additional information about this release might be needed.",
"Ideal Knowledge": "Information about LiSA's solo debut, including the release date of April 2011, the title of her mini-album 'Letters to U', and possibly details about the album's content, reception, or chart performance.",
"Key Factual Information": "mini @-@ album Letters to U",
"Prefix reformulated": "What did LiSA release for her solo debut in April 2011?",
"Prefix-intent": "In April 2011 , she made her solo debut with the release of her",
"Verification": "The KFI directly provides information about LiSA's solo debut release, as prompted by the PI."
}
]
} |
LiSA 's songs have been featured as theme music for various anime such as Fate / Zero and Sword Art Online . Her singles have regularly been in the top ten of the Oricon weekly charts , with " Crossing Field " being certified platinum by the Recording Industry Association of Japan and " Oath Sign " being certified gold . She performed at the Nippon Budokan in 2014 and 2015 . In 2015 , she made her acting debut as Madge Nelson in the Japanese dub of the animated film Minions .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n LiSA 's songs have been featured as theme music for various anime such as Fate / Zero and Sword Art Online . Her singles have regularly been in the top ten of the Oricon weekly charts , with \" Crossing Field \" being certified platinum by the Recording Industry Association of Japan and \" Oath Sign \" being certified gold . She performed at the Nippon Budokan in 2014 and 2015 . In 2015 , she made her acting debut as Madge Nelson in the Japanese dub of the animated film Minions . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>LiSA 's songs have been featured as theme music for various anime such as<|end-PI|> <|retrieval|> <|start-KFI|>Fate / Zero and Sword Art Online<|end-KFI|> . <|start-PI|>Her singles have regularly been in the top ten of the Oricon weekly charts , with<|end-PI|> <|retrieval|> <|start-KFI|>\" Crossing Field \" being certified platinum by the Recording Industry Association of Japan and \" Oath Sign \" being certified gold<|end-KFI|> . <|start-PI|>She performed at the Nippon Budokan in<|end-PI|> <|retrieval|> <|start-KFI|>2014 and 2015<|end-KFI|> . <|start-PI|>In 2015 , she made her acting debut as<|end-PI|> <|retrieval|> <|start-KFI|>Madge Nelson in the Japanese dub of the animated film Minions<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the fact that LiSA's songs have been used in anime, and the KFI provides specific examples. The <|retrieval|> token is inserted to indicate where additional information about the anime series might be needed.",
"Ideal Knowledge": "A comprehensive list of anime series that have used LiSA's songs as opening or ending themes, including details about the specific songs used and their release dates.",
"Key Factual Information": "Fate / Zero and Sword Art Online",
"Prefix reformulated": "What anime has LiSA's songs been featured as theme music for?",
"Prefix-intent": "LiSA 's songs have been featured as theme music for various anime such as",
"Verification": "The KFI directly provides examples of anime that have featured LiSA's songs as theme music, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of LiSA's success on the Oricon charts, and the KFI provides specific examples of her singles' certifications. The <|retrieval|> token is inserted to indicate where additional chart performance data might be needed.",
"Ideal Knowledge": "Detailed information about LiSA's singles' performance on the Oricon charts, including peak positions, weeks on the chart, and certification levels for each single.",
"Key Factual Information": "\" Crossing Field \" being certified platinum by the Recording Industry Association of Japan and \" Oath Sign \" being certified gold",
"Prefix reformulated": "What certifications have LiSA's singles received on the Oricon weekly charts?",
"Prefix-intent": "Her singles have regularly been in the top ten of the Oricon weekly charts , with",
"Verification": "The KFI directly provides information about the certifications of LiSA's singles, as prompted by the PI."
},
{
"Explanation": "The PI introduces LiSA's performances at the Nippon Budokan, and the KFI provides the specific years. The <|retrieval|> token is inserted to indicate where additional information about these performances might be needed.",
"Ideal Knowledge": "A list of LiSA's major concert performances, including dates, venues, and any significant milestones or achievements associated with each performance.",
"Key Factual Information": "2014 and 2015",
"Prefix reformulated": "In which years did LiSA perform at the Nippon Budokan?",
"Prefix-intent": "She performed at the Nippon Budokan in",
"Verification": "The KFI directly provides the years of LiSA's performances at the Nippon Budokan, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of LiSA's acting debut, and the KFI provides specific details about the role. The <|retrieval|> token is inserted to indicate where additional information about her acting career might be needed.",
"Ideal Knowledge": "Details about LiSA's voice acting roles, including the character of Madge Nelson, her performance in the Japanese dub of Minions, and any other voice acting work she may have done.",
"Key Factual Information": "Madge Nelson in the Japanese dub of the animated film Minions",
"Prefix reformulated": "What role did LiSA make her acting debut with in 2015?",
"Prefix-intent": "In 2015 , she made her acting debut as",
"Verification": "The KFI directly provides information about LiSA's acting debut, including the character, type of role, and film, as prompted by the PI."
}
]
} |
Oribe 's musical experience began at the age of three when , upon her mother 's suggestions , she took private piano lessons . Although she was a shy student in kindergarten , she enjoyed her music lessons . In primary school , she was inspired to sing after seeing the band Speed on television . She later took dance and vocal lessons , which continued through her junior high school years . While in elementary , she participated in an audition held at the Nippon Budokan , and it was during this time she decided to become an artist . In junior high school , she formed a band which covered songs by Avril Lavigne , Love Psychedelico , and Ego @-@ Wrappin ' .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Oribe 's musical experience began at the age of three when , upon her mother 's suggestions , she took private piano lessons . Although she was a shy student in kindergarten , she enjoyed her music lessons . In primary school , she was inspired to sing after seeing the band Speed on television . She later took dance and vocal lessons , which continued through her junior high school years . While in elementary , she participated in an audition held at the Nippon Budokan , and it was during this time she decided to become an artist . In junior high school , she formed a band which covered songs by Avril Lavigne , Love Psychedelico , and Ego @-@ Wrappin ' . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Oribe's musical experience began at the age of<|end-PI|> <|retrieval|> <|start-KFI|>three when, upon her mother's suggestions, she took private piano lessons<|end-KFI|>. Although she was a shy student in kindergarten, she enjoyed her music lessons. <|start-PI|>In primary school, she was inspired to sing after seeing<|end-PI|> <|retrieval|> <|start-KFI|>the band Speed on television<|end-KFI|>. She later took dance and vocal lessons, which continued through her junior high school years. <|start-PI|>While in elementary, she participated in an audition held at<|end-PI|> <|retrieval|> <|start-KFI|>the Nippon Budokan, and it was during this time she decided to become an artist<|end-KFI|>. <|start-PI|>In junior high school, she formed a band which covered songs by<|end-PI|> <|retrieval|> <|start-KFI|>Avril Lavigne, Love Psychedelico, and Ego-Wrappin'<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces Oribe's musical beginnings, while the KFI provides the specific age and details. The <|retrieval|> token is inserted to indicate where additional information about her early musical experiences might be needed.",
"Ideal Knowledge": "Information about Oribe's early childhood, focusing on her introduction to music, specifically mentioning her age when she started taking piano lessons and her mother's role in this decision.",
"Key Factual Information": "three when, upon her mother's suggestions, she took private piano lessons",
"Prefix reformulated": "At what age did Oribe's musical experience begin?",
"Prefix-intent": "Oribe's musical experience began at the age of",
"Verification": "The KFI directly provides the age and initial musical experience mentioned in the PI."
},
{
"Explanation": "The PI sets up the context of Oribe's inspiration to sing, and the KFI provides the specific source of inspiration. The <|retrieval|> token is inserted to indicate where additional information about the band Speed might be needed.",
"Ideal Knowledge": "Information about the band Speed, their popularity during Oribe's primary school years, and their influence on young aspiring singers in Japan.",
"Key Factual Information": "the band Speed on television",
"Prefix reformulated": "What inspired Oribe to sing during her primary school years?",
"Prefix-intent": "In primary school, she was inspired to sing after seeing",
"Verification": "The KFI directly answers what inspired Oribe to sing, as prompted by the PI."
},
{
"Explanation": "The PI introduces Oribe's participation in an audition, and the KFI provides the specific location and its impact on her career decision. The <|retrieval|> token is inserted to indicate where additional information about the Nippon Budokan might be needed.",
"Ideal Knowledge": "Information about the Nippon Budokan, its significance in the Japanese entertainment industry, and its role in hosting auditions for young talents.",
"Key Factual Information": "the Nippon Budokan, and it was during this time she decided to become an artist",
"Prefix reformulated": "Where did Oribe participate in an audition during her elementary school years?",
"Prefix-intent": "While in elementary, she participated in an audition held at",
"Verification": "The KFI directly provides the location of the audition and its impact on Oribe's career decision, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of Oribe forming a band in junior high school, and the KFI provides the specific artists whose songs they covered. The <|retrieval|> token is inserted to indicate where additional information about these artists might be needed.",
"Ideal Knowledge": "Information about Avril Lavigne, Love Psychedelico, and Ego-Wrappin', their musical styles, popularity in Japan during Oribe's junior high school years, and their influence on young musicians.",
"Key Factual Information": "Avril Lavigne, Love Psychedelico, and Ego-Wrappin'",
"Prefix reformulated": "Which artists' songs did Oribe's band cover in junior high school?",
"Prefix-intent": "In junior high school, she formed a band which covered songs by",
"Verification": "The KFI directly lists the artists whose songs were covered by Oribe's band, as prompted by the PI."
}
]
} |
Oribe began her singing career in 2005 during her first year in high school when she formed the indie rock band Chucky , which mainly covered songs by other artists . During her second year , the band received advice from their peers that they should start making their own songs ; eventually the band 's reputation grew enough that they performed in Osaka and Nagoya in addition to Gifu . Although her grades in high school were good , she decided not to go to university , despite the recommendations of her teachers , in order to focus on her work with Chucky . Nevertheless , after graduation , due to varying schedules among the band 's members , it became difficult to continue performing . Following the band 's disbandment in July 2008 , she moved to Tokyo in order to continue her singing career .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Oribe began her singing career in 2005 during her first year in high school when she formed the indie rock band Chucky , which mainly covered songs by other artists . During her second year , the band received advice from their peers that they should start making their own songs ; eventually the band 's reputation grew enough that they performed in Osaka and Nagoya in addition to Gifu . Although her grades in high school were good , she decided not to go to university , despite the recommendations of her teachers , in order to focus on her work with Chucky . Nevertheless , after graduation , due to varying schedules among the band 's members , it became difficult to continue performing . Following the band 's disbandment in July 2008 , she moved to Tokyo in order to continue her singing career . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Oribe began her singing career in 2005 during her first year in high school when she<|end-PI|> <|retrieval|> <|start-KFI|>formed the indie rock band Chucky , which mainly covered songs by other artists<|end-KFI|> . <|start-PI|>During her second year , the band received advice from their peers that they should<|end-PI|> <|retrieval|> <|start-KFI|>start making their own songs ; eventually the band 's reputation grew enough that they performed in Osaka and Nagoya in addition to Gifu<|end-KFI|> . <|start-PI|>Although her grades in high school were good , she decided not to go to university , despite the recommendations of her teachers , in order to<|end-PI|> <|retrieval|> <|start-KFI|>focus on her work with Chucky<|end-KFI|> . Nevertheless , after graduation , due to varying schedules among the band 's members , it became difficult to continue performing . <|start-PI|>Following the band 's disbandment in July 2008 , she moved to Tokyo in order to<|end-PI|> <|retrieval|> <|start-KFI|>continue her singing career<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces Oribe's career start, and the KFI provides specific details about her first band. The <|retrieval|> token is inserted to indicate where additional information about her career beginning might be needed.",
"Ideal Knowledge": "Details about Oribe's early musical career, including the formation of her first band Chucky, its genre, and initial repertoire.",
"Key Factual Information": "formed the indie rock band Chucky , which mainly covered songs by other artists",
"Prefix reformulated": "How did Oribe begin her singing career in 2005?",
"Prefix-intent": "Oribe began her singing career in 2005 during her first year in high school when she",
"Verification": "The KFI directly provides information about how Oribe started her singing career, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context for advice received, and the KFI provides the specific advice and its results. The <|retrieval|> token is inserted to indicate where additional information about the band's development might be needed.",
"Ideal Knowledge": "Information about the band's development, including the transition to original compositions and expansion of their performance venues.",
"Key Factual Information": "start making their own songs ; eventually the band 's reputation grew enough that they performed in Osaka and Nagoya in addition to Gifu",
"Prefix reformulated": "What advice did Oribe's band receive during her second year?",
"Prefix-intent": "During her second year , the band received advice from their peers that they should",
"Verification": "The KFI directly provides the advice received and its consequences, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of Oribe's academic performance and decision, and the KFI provides the specific reason for her choice. The <|retrieval|> token is inserted to indicate where additional information about her career decisions might be needed.",
"Ideal Knowledge": "Information about Oribe's career priorities and decision-making process regarding her education and music career.",
"Key Factual Information": "focus on her work with Chucky",
"Prefix reformulated": "Why did Oribe decide not to go to university despite good grades?",
"Prefix-intent": "Although her grades in high school were good , she decided not to go to university , despite the recommendations of her teachers , in order to",
"Verification": "The KFI directly provides the reason for Oribe's decision, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of the band's disbandment and Oribe's move, and the KFI provides the specific reason for her relocation. The <|retrieval|> token is inserted to indicate where additional information about her career decisions might be needed.",
"Ideal Knowledge": "Information about Oribe's career progression, including her relocation to Tokyo and her motivations for continuing her singing career.",
"Key Factual Information": "continue her singing career",
"Prefix reformulated": "Why did Oribe move to Tokyo after her band disbanded in July 2008?",
"Prefix-intent": "Following the band 's disbandment in July 2008 , she moved to Tokyo in order to",
"Verification": "The KFI directly provides the reason for Oribe's move to Tokyo, as prompted by the PI."
}
]
} |
After moving to Tokyo , Oribe formed the band Love is Same All with members from the indie band Parking Out and began using the stage name LiSA , which is an acronym for Love is Same All . The band performs with LiSA during the latter 's solo live performances . In 2010 , she made her major debut singing songs for the anime series Angel Beats ! as one of two vocalists for the fictional in @-@ story band Girls Dead Monster . She was the vocalist for the character Yui , and the second vocalist , Marina , sang as the character Masami Iwasawa . LiSA put out three singles and one album in 2010 under the name Girls Dead Monster on Key 's record label Key Sounds Label . The first single " Thousand Enemies " was released on May 12 ; the second single " Little Braver " came out on June 9 ; and the third single " Ichiban no Takaramono ( Yui final ver . ) " ( 一番の宝物 〜 Yui final ver . 〜 , " My Most Precious Treasure ( Yui final ver . ) " ) was sold on December 8 . The album Keep The Beats ! was released on June 30 . LiSA made her first appearance at Animelo Summer Live during the concert 's 2010 iteration on August 28 .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n After moving to Tokyo , Oribe formed the band Love is Same All with members from the indie band Parking Out and began using the stage name LiSA , which is an acronym for Love is Same All . The band performs with LiSA during the latter 's solo live performances . In 2010 , she made her major debut singing songs for the anime series Angel Beats ! as one of two vocalists for the fictional in @-@ story band Girls Dead Monster . She was the vocalist for the character Yui , and the second vocalist , Marina , sang as the character Masami Iwasawa . LiSA put out three singles and one album in 2010 under the name Girls Dead Monster on Key 's record label Key Sounds Label . The first single \" Thousand Enemies \" was released on May 12 ; the second single \" Little Braver \" came out on June 9 ; and the third single \" Ichiban no Takaramono ( Yui final ver . ) \" ( 一番の宝物 〜 Yui final ver . 〜 , \" My Most Precious Treasure ( Yui final ver . ) \" ) was sold on December 8 . The album Keep The Beats ! was released on June 30 . LiSA made her first appearance at Animelo Summer Live during the concert 's 2010 iteration on August 28 . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>After moving to Tokyo, Oribe formed the band Love is Same All with members from the indie band Parking Out and began using the stage name<|end-PI|> <|retrieval|> <|start-KFI|>LiSA, which is an acronym for Love is Same All<|end-KFI|>. The band performs with LiSA during the latter's solo live performances. <|start-PI|>In 2010, she made her major debut singing songs for the anime series<|end-PI|> <|retrieval|> <|start-KFI|>Angel Beats! as one of two vocalists for the fictional in-story band Girls Dead Monster<|end-KFI|>. She was the vocalist for the character Yui, and the second vocalist, Marina, sang as the character Masami Iwasawa. <|start-PI|>LiSA put out three singles and one album in 2010 under the name Girls Dead Monster on Key's record label<|end-PI|> <|retrieval|> <|start-KFI|>Key Sounds Label<|end-KFI|>. The first single \"Thousand Enemies\" was released on May 12; the second single \"Little Braver\" came out on June 9; and the third single \"Ichiban no Takaramono (Yui final ver.)\" (一番の宝物 〜 Yui final ver. 〜, \"My Most Precious Treasure (Yui final ver.)\") was sold on December 8. The album Keep The Beats! was released on June 30. <|start-PI|>LiSA made her first appearance at Animelo Summer Live during the concert's<|end-PI|> <|retrieval|> <|start-KFI|>2010 iteration on August 28<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context for Oribe's new stage name, and the KFI provides the specific name and its meaning. The <|retrieval|> token is inserted to indicate where additional information about the artist's name change might be needed.",
"Ideal Knowledge": "Information about Japanese singer LiSA, including her real name (Oribe), the formation of her band Love is Same All, and the origin of her stage name.",
"Key Factual Information": "LiSA, which is an acronym for Love is Same All",
"Prefix reformulated": "What stage name did Oribe begin using after forming the band Love is Same All?",
"Prefix-intent": "After moving to Tokyo, Oribe formed the band Love is Same All with members from the indie band Parking Out and began using the stage name",
"Verification": "The KFI directly provides the stage name and its meaning, as prompted by the PI."
},
{
"Explanation": "The PI introduces LiSA's major debut, and the KFI specifies the anime series and her role. The <|retrieval|> token is inserted to indicate where additional information about the anime and LiSA's involvement might be needed.",
"Ideal Knowledge": "Detailed information about the anime series Angel Beats!, its soundtrack, and the fictional band Girls Dead Monster, including LiSA's role as one of the vocalists.",
"Key Factual Information": "Angel Beats! as one of two vocalists for the fictional in-story band Girls Dead Monster",
"Prefix reformulated": "What anime series did LiSA make her major debut singing songs for in 2010?",
"Prefix-intent": "In 2010, she made her major debut singing songs for the anime series",
"Verification": "The KFI directly provides the anime series and the specific role LiSA played, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context for LiSA's releases as Girls Dead Monster, and the KFI provides the specific record label. The <|retrieval|> token is inserted to indicate where additional information about the label might be needed.",
"Ideal Knowledge": "Information about Key Sounds Label, its association with the visual novel company Key, and its role in producing music for anime adaptations of Key's works, including Angel Beats!",
"Key Factual Information": "Key Sounds Label",
"Prefix reformulated": "What record label did LiSA release her music under as Girls Dead Monster in 2010?",
"Prefix-intent": "LiSA put out three singles and one album in 2010 under the name Girls Dead Monster on Key's record label",
"Verification": "The KFI directly provides the name of the record label, as prompted by the PI."
},
{
"Explanation": "The PI introduces LiSA's first appearance at Animelo Summer Live, and the KFI specifies the exact date and year. The <|retrieval|> token is inserted to indicate where additional information about the concert and LiSA's participation might be needed.",
"Ideal Knowledge": "Information about Animelo Summer Live, its significance in the anime music industry, and details about the 2010 event, including the lineup and LiSA's performance.",
"Key Factual Information": "2010 iteration on August 28",
"Prefix reformulated": "When did LiSA make her first appearance at Animelo Summer Live?",
"Prefix-intent": "LiSA made her first appearance at Animelo Summer Live during the concert's",
"Verification": "The KFI directly provides the year and date of LiSA's first appearance at the concert, as prompted by the PI."
}
]
} |
LiSA made her solo debut on April 20 , 2011 with the release of her mini @-@ album Letters to U by Aniplex under Sony Music Artists . The songs on the album were composed by dōjin and major artists , and she composed the first song " Believe in Myself " ; she wrote the album 's lyrics . On November 12 , 2011 , She made her first appearance in Singapore at Anime Festival Asia , Southeast Asia 's largest anime and pop culture convention . She released her first solo single " Oath Sign " on November 23 , 2011 , which was used as the opening theme to the 2011 anime series Fate / Zero . The single peaked at No. 5 on the Oricon weekly charts and was certified gold by the Recording Industry Association of Japan ( RIAJ ) .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n LiSA made her solo debut on April 20 , 2011 with the release of her mini @-@ album Letters to U by Aniplex under Sony Music Artists . The songs on the album were composed by dōjin and major artists , and she composed the first song \" Believe in Myself \" ; she wrote the album 's lyrics . On November 12 , 2011 , She made her first appearance in Singapore at Anime Festival Asia , Southeast Asia 's largest anime and pop culture convention . She released her first solo single \" Oath Sign \" on November 23 , 2011 , which was used as the opening theme to the 2011 anime series Fate / Zero . The single peaked at No. 5 on the Oricon weekly charts and was certified gold by the Recording Industry Association of Japan ( RIAJ ) . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>LiSA made her solo debut on April 20 , 2011 with the release of<|end-PI|> <|retrieval|> <|start-KFI|>her mini @-@ album Letters to U by Aniplex under Sony Music Artists<|end-KFI|> . The songs on the album were composed by dōjin and major artists , and she composed the first song \" Believe in Myself \" ; she wrote the album 's lyrics . <|start-PI|>On November 12 , 2011 , She made her first appearance in Singapore at<|end-PI|> <|retrieval|> <|start-KFI|>Anime Festival Asia , Southeast Asia 's largest anime and pop culture convention<|end-KFI|> . <|start-PI|>She released her first solo single \" Oath Sign \" on November 23 , 2011 , which was used as<|end-PI|> <|retrieval|> <|start-KFI|>the opening theme to the 2011 anime series Fate / Zero<|end-KFI|> . The single peaked at No. 5 on the Oricon weekly charts and was certified gold by the Recording Industry Association of Japan ( RIAJ ) .",
"none": null,
"results": [
{
"Explanation": "The PI introduces LiSA's solo debut and sets up a query about her release, while the KFI provides specific details about the album. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "Detailed information about LiSA's discography, including her debut release 'Letters to U', its format, release date, and the record labels involved.",
"Key Factual Information": "her mini @-@ album Letters to U by Aniplex under Sony Music Artists",
"Prefix reformulated": "What did LiSA release for her solo debut on April 20, 2011?",
"Prefix-intent": "LiSA made her solo debut on April 20 , 2011 with the release of",
"Verification": "The KFI directly provides the information about LiSA's debut release as requested in the PI."
},
{
"Explanation": "The PI sets up the context for LiSA's first appearance in Singapore, and the KFI provides specific details about the event. The <|retrieval|> token is inserted to indicate where additional information about the event might be needed.",
"Ideal Knowledge": "Information about Anime Festival Asia, including its significance in Southeast Asia, its annual schedule, and notable performers or guests, particularly focusing on LiSA's appearance in 2011.",
"Key Factual Information": "Anime Festival Asia , Southeast Asia 's largest anime and pop culture convention",
"Prefix reformulated": "Where did LiSA make her first appearance in Singapore on November 12, 2011?",
"Prefix-intent": "On November 12 , 2011 , She made her first appearance in Singapore at",
"Verification": "The KFI directly provides the information about the event where LiSA made her first appearance in Singapore, as prompted by the PI."
},
{
"Explanation": "The PI introduces LiSA's first solo single and sets up a query about its use, while the KFI provides the specific information about its role in an anime series. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "Details about LiSA's single 'Oath Sign', including its release date, its use in the anime Fate/Zero, and information about the anime series itself.",
"Key Factual Information": "the opening theme to the 2011 anime series Fate / Zero",
"Prefix reformulated": "What was LiSA's first solo single 'Oath Sign' used for when it was released on November 23, 2011?",
"Prefix-intent": "She released her first solo single \" Oath Sign \" on November 23 , 2011 , which was used as",
"Verification": "The KFI directly provides the information about the use of LiSA's single as requested in the PI."
}
]
} |
LiSA released her first full solo album Lover " s " mile on February 22 , 2012 ; the album peaked at No. 7 on the Oricon weekly charts . She attended the Anime Expo 2012 in Los Angeles as a guest of honor and performed her first concert in North America there on July 1 . After her success with " Oath Sign " , she was chosen to perform the first opening theme to the 2012 anime series Sword Art Online ; the single " Crossing Field " , her second single , was released on August 8 , 2012 ; the single peaked at No. 5 on Oricon and was later certified gold by the RIAJ . Her third single " Best Day , Best Way " , which peaked at No. 6 on Oricon , was released on April 3 , 2013 , and her fourth single " Träumerei " , which peaked at No. 15 on Oricon and was used as the opening theme to the 2013 anime series Day Break Illusion , was released on August 7 , 2013 . LiSA released her second solo album Landspace on October 30 , 2013 ; the album peaked at No. 2 on Oricon charts . A photobook titled Kyō mo Ii Hi da ( 今日もいい日だっ , " Another Great Day " ) that contains photos of her taken by Lis Ani ! magazine over three years was published on November 22 , 2013 . When asked if she would revisit the meaning of her acronym in the future in an interview with Nihongogo at Anime Festival Asia 2013 in Singapore , she suggested Lovely international Super Apple .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n LiSA released her first full solo album Lover \" s \" mile on February 22 , 2012 ; the album peaked at No. 7 on the Oricon weekly charts . She attended the Anime Expo 2012 in Los Angeles as a guest of honor and performed her first concert in North America there on July 1 . After her success with \" Oath Sign \" , she was chosen to perform the first opening theme to the 2012 anime series Sword Art Online ; the single \" Crossing Field \" , her second single , was released on August 8 , 2012 ; the single peaked at No. 5 on Oricon and was later certified gold by the RIAJ . Her third single \" Best Day , Best Way \" , which peaked at No. 6 on Oricon , was released on April 3 , 2013 , and her fourth single \" Träumerei \" , which peaked at No. 15 on Oricon and was used as the opening theme to the 2013 anime series Day Break Illusion , was released on August 7 , 2013 . LiSA released her second solo album Landspace on October 30 , 2013 ; the album peaked at No. 2 on Oricon charts . A photobook titled Kyō mo Ii Hi da ( 今日もいい日だっ , \" Another Great Day \" ) that contains photos of her taken by Lis Ani ! magazine over three years was published on November 22 , 2013 . When asked if she would revisit the meaning of her acronym in the future in an interview with Nihongogo at Anime Festival Asia 2013 in Singapore , she suggested Lovely international Super Apple . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
On January 3 , 2014 , LiSA held a sold @-@ out solo concert at the Nippon Budokan . Her fifth single " Rising Hope " , which is used as the opening theme to the 2014 anime series The Irregular at Magic High School , was released on May 7 , 2014 ; the single peaked at No. 4 on Oricon . She later covered the songs " Headphone Actor " ( ヘッドフォンアクター ) and " Yūkei Yesterday " ( 夕景イエスタデイ , " Yesterday Evening " ) from the Kagerou Project franchise for the sixth episode of the 2014 anime series Mekakucity Actors . She released her sixth single " Bright Flight / L. Miranic " on September 17 , 2014 , which peaked at No. 8 on Oricon , and her seventh single " Shirushi " ( シルシ , lit . " Sign " ) on December 10 , 2014 , which peaked at No. 3 on the Oricon ; the title song is used as the third ending theme to the 2014 anime series Sword Art Online II , and the single also includes the song " No More Time Machine " , which was used as the second ending theme to Sword Art Online II . She held her second Nippon Budokan concert on January 10 and 11 , 2015 , tickets for which were sold out . She released her third solo album Launcher on March 4 , 2015 and her eighth single " Rally Go Round " on May 27 , 2015 ; the song is used as the opening theme to the second season of the anime television series Nisekoi . She was cast as Madge Nelson in the Japanese dub of the animated film Minions , which premiered in Japanese theaters on July 31 , 2015 . She released her ninth single " Empty Mermaid " on September 30 , 2015 . In late 2015 , she performed " ID " , which is used as theme song of the 2015 video game Dengeki Bunko : Fighting Climax Ignition . To commemorate her fifth year as a solo artist , LiSA released her Letters To U EP as a limited edition LP on March 23 , 2016 . LiSA released a mini @-@ album titled Lucky Hi Five ! on April 20 , 2016 . She will release the single " Brave Freak Out " , which will be used as the opening theme to the 2016 anime television series Qualidea Code , on August 24 , 2016 .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n On January 3 , 2014 , LiSA held a sold @-@ out solo concert at the Nippon Budokan . Her fifth single \" Rising Hope \" , which is used as the opening theme to the 2014 anime series The Irregular at Magic High School , was released on May 7 , 2014 ; the single peaked at No. 4 on Oricon . She later covered the songs \" Headphone Actor \" ( ヘッドフォンアクター ) and \" Yūkei Yesterday \" ( 夕景イエスタデイ , \" Yesterday Evening \" ) from the Kagerou Project franchise for the sixth episode of the 2014 anime series Mekakucity Actors . She released her sixth single \" Bright Flight / L. Miranic \" on September 17 , 2014 , which peaked at No. 8 on Oricon , and her seventh single \" Shirushi \" ( シルシ , lit . \" Sign \" ) on December 10 , 2014 , which peaked at No. 3 on the Oricon ; the title song is used as the third ending theme to the 2014 anime series Sword Art Online II , and the single also includes the song \" No More Time Machine \" , which was used as the second ending theme to Sword Art Online II . She held her second Nippon Budokan concert on January 10 and 11 , 2015 , tickets for which were sold out . She released her third solo album Launcher on March 4 , 2015 and her eighth single \" Rally Go Round \" on May 27 , 2015 ; the song is used as the opening theme to the second season of the anime television series Nisekoi . She was cast as Madge Nelson in the Japanese dub of the animated film Minions , which premiered in Japanese theaters on July 31 , 2015 . She released her ninth single \" Empty Mermaid \" on September 30 , 2015 . In late 2015 , she performed \" ID \" , which is used as theme song of the 2015 video game Dengeki Bunko : Fighting Climax Ignition . To commemorate her fifth year as a solo artist , LiSA released her Letters To U EP as a limited edition LP on March 23 , 2016 . LiSA released a mini @-@ album titled Lucky Hi Five ! on April 20 , 2016 . She will release the single \" Brave Freak Out \" , which will be used as the opening theme to the 2016 anime television series Qualidea Code , on August 24 , 2016 . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
LiSA lists Avril Lavigne , Oasis , Green Day , Paramore , Ke $ ha , and Rihanna as among her musical influences , as well as her time in Chucky . LiSA wrote the lyrics for some of her songs in her Landspace and Launcher albums , as well as the lyrics for the singles " Bright Flight / L. Miranic " , " Shirushi " , and " Rally Go Round " ; " Rally Go Round " was co @-@ written with songwriter Shin Furuya .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n LiSA lists Avril Lavigne , Oasis , Green Day , Paramore , Ke $ ha , and Rihanna as among her musical influences , as well as her time in Chucky . LiSA wrote the lyrics for some of her songs in her Landspace and Launcher albums , as well as the lyrics for the singles \" Bright Flight / L. Miranic \" , \" Shirushi \" , and \" Rally Go Round \" ; \" Rally Go Round \" was co @-@ written with songwriter Shin Furuya . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>LiSA lists Avril Lavigne , Oasis , Green Day , Paramore , Ke $ ha , and Rihanna as among her<|end-PI|> <|retrieval|> <|start-KFI|>musical influences , as well as her time in Chucky<|end-KFI|>. <|start-PI|>LiSA wrote the lyrics for some of her songs in her<|end-PI|> <|retrieval|> <|start-KFI|>Landspace and Launcher albums<|end-KFI|> , as well as <|start-PI|>LiSA wrote the lyrics for the singles<|end-PI|> <|retrieval|> <|start-KFI|>\" Bright Flight / L. Miranic \" , \" Shirushi \" , and \" Rally Go Round \"<|end-KFI|> ; <|start-PI|>\" Rally Go Round \" was<|end-PI|> <|retrieval|> <|start-KFI|>co @-@ written with songwriter Shin Furuya<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces LiSA and sets up a query about her musical influences, while the KFI provides the specific artists and additional information. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "A comprehensive list of LiSA's musical influences, including both Western and Japanese artists, as well as her involvement in previous musical projects or bands.",
"Key Factual Information": "musical influences , as well as her time in Chucky",
"Prefix reformulated": "What are LiSA's musical influences?",
"Prefix-intent": "LiSA lists Avril Lavigne , Oasis , Green Day , Paramore , Ke $ ha , and Rihanna as among her",
"Verification": "The KFI directly provides the musical influences mentioned in the PI."
},
{
"Explanation": "The PI sets up a query about LiSA's lyric-writing contributions, and the KFI provides the specific album names. The <|retrieval|> token is inserted to indicate where additional information about her songwriting might be needed.",
"Ideal Knowledge": "A detailed discography of LiSA, highlighting the albums 'Landspace' and 'Launcher', with information on which songs from these albums have lyrics written by LiSA herself.",
"Key Factual Information": "Landspace and Launcher albums",
"Prefix reformulated": "Which albums contain songs with lyrics written by LiSA?",
"Prefix-intent": "LiSA wrote the lyrics for some of her songs in her",
"Verification": "The KFI directly provides the album names for which LiSA wrote lyrics, as prompted by the PI."
},
{
"Explanation": "The PI introduces LiSA's lyric-writing for singles, and the KFI provides the specific song titles. The <|retrieval|> token is inserted to indicate where additional information about these singles might be needed.",
"Ideal Knowledge": "A comprehensive list of singles released by LiSA, with details on her songwriting contributions, particularly focusing on 'Bright Flight / L. Miranic', 'Shirushi', and 'Rally Go Round'.",
"Key Factual Information": "\" Bright Flight / L. Miranic \" , \" Shirushi \" , and \" Rally Go Round \"",
"Prefix reformulated": "Which singles have lyrics written by LiSA?",
"Prefix-intent": "LiSA wrote the lyrics for the singles",
"Verification": "The KFI directly lists the singles with lyrics written by LiSA, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the writing of 'Rally Go Round', and the KFI provides information about the co-writer. The <|retrieval|> token is inserted to indicate where additional information about the song's creation might be needed.",
"Ideal Knowledge": "Detailed information about the songwriting process for 'Rally Go Round', including the collaboration between LiSA and Shin Furuya, their roles in the writing process, and any background on their working relationship.",
"Key Factual Information": "co @-@ written with songwriter Shin Furuya",
"Prefix reformulated": "How was the song 'Rally Go Round' written?",
"Prefix-intent": "\" Rally Go Round \" was",
"Verification": "The KFI directly provides information about the co-writing of 'Rally Go Round', as prompted by the PI."
}
]
} |
LiSA 's musical style is described by Dennis Amith of J ! -ENT as a young woman with style , beautiful vocals , and the ability to take on various musical styles , may it be happy , upbeat rock music or " even something more darker . " She has used the phrase " kyō mo ii hi da " ( 今日もいい日だっ , " Another Great Day " ) as a theme throughout her career ; the phrase is also the name of her personal blog . In an interview with HMV Japan , she mentioned that before her major debut , she tended to sing pop songs , and because of this , she felt uneasy when making " Oath Sign " , which had a heavy and dark style , although she was able to receive several favorable responses to the single , which allowed her to make her next single " Crossing Field " with confidence . As for " Crossing Field " , she says that feels that the song grows every time she sings it , in the same way that love is nurtured through dates or love letters . In " Best Day , Best Way " , she mentioned that the song 's theme is " what you did yesterday will lead to what happens today " , and that the lyrics are delivered as if they are her own words . She also shared her experience with the single 's B @-@ sides : " I 'm a Rock Star " , which according to her represents her fulfilling her dream of becoming a singer , and " Shiroi Toiki " ( シロイトイキ , " White Sigh " ) , which represents happy feelings .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n LiSA 's musical style is described by Dennis Amith of J ! -ENT as a young woman with style , beautiful vocals , and the ability to take on various musical styles , may it be happy , upbeat rock music or \" even something more darker . \" She has used the phrase \" kyō mo ii hi da \" ( 今日もいい日だっ , \" Another Great Day \" ) as a theme throughout her career ; the phrase is also the name of her personal blog . In an interview with HMV Japan , she mentioned that before her major debut , she tended to sing pop songs , and because of this , she felt uneasy when making \" Oath Sign \" , which had a heavy and dark style , although she was able to receive several favorable responses to the single , which allowed her to make her next single \" Crossing Field \" with confidence . As for \" Crossing Field \" , she says that feels that the song grows every time she sings it , in the same way that love is nurtured through dates or love letters . In \" Best Day , Best Way \" , she mentioned that the song 's theme is \" what you did yesterday will lead to what happens today \" , and that the lyrics are delivered as if they are her own words . She also shared her experience with the single 's B @-@ sides : \" I 'm a Rock Star \" , which according to her represents her fulfilling her dream of becoming a singer , and \" Shiroi Toiki \" ( シロイトイキ , \" White Sigh \" ) , which represents happy feelings . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
In an interview with Oricon , LiSA explained her experiences with " Träumerei " , where she stated that because Day Break Illusion is an original anime , she wanted the song to be presented in the color of her own work , but even in the world of anime , her music would continue to have an atmosphere of rock , as opposed to the pop style she used in " Best Day , Best Way " . She mentioned that she sang that song in a way that she felt the conflicts that were present in the story and with the strength that was part of the show 's theme . With regards to the song 's music video , she wanted to have a video which would express in color emotions , such as the use of red to represent confrontations , black to represent the feeling of being lost , and green for frightening things .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n In an interview with Oricon , LiSA explained her experiences with \" Träumerei \" , where she stated that because Day Break Illusion is an original anime , she wanted the song to be presented in the color of her own work , but even in the world of anime , her music would continue to have an atmosphere of rock , as opposed to the pop style she used in \" Best Day , Best Way \" . She mentioned that she sang that song in a way that she felt the conflicts that were present in the story and with the strength that was part of the show 's theme . With regards to the song 's music video , she wanted to have a video which would express in color emotions , such as the use of red to represent confrontations , black to represent the feeling of being lost , and green for frightening things . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
In making the single " Bright Flight / L. Miranic " , she mentions in an interview with HMV Japan that the concept of the single is that it can be divided into two parts : pink and black . Pink is the color of " Bright Flight " , a song which is in a pop style and was written in a " cute " image in mind , while black is the color of " L. Miranic " , which is in a dark rock style and has the theme of a " bad woman " . LiSA mentioned that the name " L. Miranic " was chosen because the song is linked to criminals and that since " Miranic " is a name in some countries , adding her name to " Miranic " to make " LiSA @-@ Miranic " would mean that the criminal in the song is her .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n In making the single \" Bright Flight / L. Miranic \" , she mentions in an interview with HMV Japan that the concept of the single is that it can be divided into two parts : pink and black . Pink is the color of \" Bright Flight \" , a song which is in a pop style and was written in a \" cute \" image in mind , while black is the color of \" L. Miranic \" , which is in a dark rock style and has the theme of a \" bad woman \" . LiSA mentioned that the name \" L. Miranic \" was chosen because the song is linked to criminals and that since \" Miranic \" is a name in some countries , adding her name to \" Miranic \" to make \" LiSA @-@ Miranic \" would mean that the criminal in the song is her . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
Dennis Amith of J ! -ENT reviewed LiSA 's album Landspace , where he describes the song " Crossing Field " as a song about " wanting to forget the weakness and securities of one past and being with someone you love who gives you the strength to become even stronger . " Meanwhile , he describes the song " Best Day , Best Way " as " a fun and happy track about believing in yourself and overcoming anything bad that have happened in the past and moving forward . " Finally , he describes " Träumerei " as " another inspirational song about one losing the rhythm in their heart and is now on the road of loneliness , but still the person wants to pierce the sky with their own light , shatter the darkness and have a fresh new start in life . " He concludes the review by saying that the album is full of upbeat and inspirational songs and showcases LiSA 's talents as a vocalist , and in contrast to many anison artists who are known for their cute and sweet vocals , LiSA is able to take on many musical styles .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Dennis Amith of J ! -ENT reviewed LiSA 's album Landspace , where he describes the song \" Crossing Field \" as a song about \" wanting to forget the weakness and securities of one past and being with someone you love who gives you the strength to become even stronger . \" Meanwhile , he describes the song \" Best Day , Best Way \" as \" a fun and happy track about believing in yourself and overcoming anything bad that have happened in the past and moving forward . \" Finally , he describes \" Träumerei \" as \" another inspirational song about one losing the rhythm in their heart and is now on the road of loneliness , but still the person wants to pierce the sky with their own light , shatter the darkness and have a fresh new start in life . \" He concludes the review by saying that the album is full of upbeat and inspirational songs and showcases LiSA 's talents as a vocalist , and in contrast to many anison artists who are known for their cute and sweet vocals , LiSA is able to take on many musical styles . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
The discography of LiSA includes three studio albums , one extended play , ten singles , and five video albums .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n The discography of LiSA includes three studio albums , one extended play , ten singles , and five video albums . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>The discography of LiSA includes<|end-PI|> <|retrieval|> <|start-KFI|>three studio albums , one extended play , ten singles , and five video albums<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces LiSA's discography and sets up a query about its contents, while the KFI provides the specific details of her musical releases. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "A comprehensive list of LiSA's musical releases, categorized by type (studio albums, EPs, singles, video albums) with their respective counts.",
"Key Factual Information": "three studio albums , one extended play , ten singles , and five video albums",
"Prefix reformulated": "What does the discography of LiSA include?",
"Prefix-intent": "The discography of LiSA includes",
"Verification": "The KFI directly lists the components of LiSA's discography, as prompted by the PI."
}
]
} |
Aston Villa Football Club ( / ˈæstən ˈvɪlə / ; nicknamed Villa , The Villa , The Villans , The Lions ) is a professional association football club based in Aston , Birmingham , that plays in the Championship , the second level of English football . Founded in 1874 , they have played at their current home ground , Villa Park , since 1897 . Aston Villa were the originators and founding members of the Football League in 1888 . They were also founding members of the Premier League in 1992 . In June 2016 , the club was sold by American businessman Randy Lerner to Recon Group , owned by Chinese businessman Dr Tony Jiantong Xia .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Aston Villa Football Club ( / ˈæstən ˈvɪlə / ; nicknamed Villa , The Villa , The Villans , The Lions ) is a professional association football club based in Aston , Birmingham , that plays in the Championship , the second level of English football . Founded in 1874 , they have played at their current home ground , Villa Park , since 1897 . Aston Villa were the originators and founding members of the Football League in 1888 . They were also founding members of the Premier League in 1992 . In June 2016 , the club was sold by American businessman Randy Lerner to Recon Group , owned by Chinese businessman Dr Tony Jiantong Xia . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
Aston Villa are one of the oldest and most successful football clubs in the history of English football . Villa won the 1981 – 82 European Cup , and are thus one of five English clubs to win what is now the UEFA Champions League . They have the fifth highest total of major honours won by an English club , having won the First Division Championship seven times , the FA Cup seven times , the Football League Cup five times , and the European Cup and UEFA Super Cup double in 1982 .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Aston Villa are one of the oldest and most successful football clubs in the history of English football . Villa won the 1981 – 82 European Cup , and are thus one of five English clubs to win what is now the UEFA Champions League . They have the fifth highest total of major honours won by an English club , having won the First Division Championship seven times , the FA Cup seven times , the Football League Cup five times , and the European Cup and UEFA Super Cup double in 1982 . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
They have a fierce local rivalry with Birmingham City . The Second City derby between Aston Villa and Birmingham City has been played since 1879 . The club 's traditional kit colours are claret shirts with sky blue sleeves , white shorts and sky blue socks . Their traditional badge is of a rampant lion , which was introduced by the club 's Scottish chairman William McGregor in honour of the Royal Standard of Scotland .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n They have a fierce local rivalry with Birmingham City . The Second City derby between Aston Villa and Birmingham City has been played since 1879 . The club 's traditional kit colours are claret shirts with sky blue sleeves , white shorts and sky blue socks . Their traditional badge is of a rampant lion , which was introduced by the club 's Scottish chairman William McGregor in honour of the Royal Standard of Scotland . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
Aston Villa Football Club were formed in March 1874 , by members of the Villa Cross Wesleyan Chapel in Handsworth which is now part of Birmingham . The four founders of Aston Villa were Jack Hughes , Frederick Matthews , Walter Price and William Scattergood . Aston Villa 's first match was against the local Aston Brook St Mary 's Rugby team . As a condition of the match , the Villa side had to agree to play the first half under rugby rules and the second half under football rules . After moving to the Wellington Road ground in 1876 , Villa soon established themselves as one of the best teams in the Midlands , winning their first honour , the Birmingham Senior Cup in 1880 , under the captaincy of Scotsman George Ramsay .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Aston Villa Football Club were formed in March 1874 , by members of the Villa Cross Wesleyan Chapel in Handsworth which is now part of Birmingham . The four founders of Aston Villa were Jack Hughes , Frederick Matthews , Walter Price and William Scattergood . Aston Villa 's first match was against the local Aston Brook St Mary 's Rugby team . As a condition of the match , the Villa side had to agree to play the first half under rugby rules and the second half under football rules . After moving to the Wellington Road ground in 1876 , Villa soon established themselves as one of the best teams in the Midlands , winning their first honour , the Birmingham Senior Cup in 1880 , under the captaincy of Scotsman George Ramsay . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
The club won their first FA Cup in 1887 with captain Archie Hunter becoming one of the game 's first household names . Aston Villa were one of the dozen teams that competed in the inaugural Football League in 1888 with one of the club 's directors , William McGregor being the league 's founder . Aston Villa emerged as the most successful English club of the Victorian era , winning no fewer than five League titles and three FA Cups by the end of Queen Victoria 's reign . In 1897 , the year Villa won The Double , they moved into their present home , the Aston Lower Grounds . Supporters coined the name " Villa Park " ; no official declaration listed the ground as Villa Park .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n The club won their first FA Cup in 1887 with captain Archie Hunter becoming one of the game 's first household names . Aston Villa were one of the dozen teams that competed in the inaugural Football League in 1888 with one of the club 's directors , William McGregor being the league 's founder . Aston Villa emerged as the most successful English club of the Victorian era , winning no fewer than five League titles and three FA Cups by the end of Queen Victoria 's reign . In 1897 , the year Villa won The Double , they moved into their present home , the Aston Lower Grounds . Supporters coined the name \" Villa Park \" ; no official declaration listed the ground as Villa Park . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
Aston Villa won their sixth FA Cup in 1920 , soon after though the club began a slow decline that led to Villa , at the time one of the most famous and successful clubs in world football , being relegated in 1936 for the first time to the Second Division . This was largely the result of a dismal defensive record : they conceded 110 goals in 42 games , 7 of them coming from Arsenal 's Ted Drake in an infamous 1 – 7 defeat at Villa Park . Like all English clubs , Villa lost seven seasons to the Second World War , and that conflict brought several careers to a premature end . The team was rebuilt under the guidance of former player Alex Massie for the remainder of the 1940s . Aston Villa 's first trophy for 37 years came in the 1956 – 57 season when another former Villa player , Eric Houghton led the club to a then record seventh FA Cup Final win , defeating the ' Busby Babes ' of Manchester United in the final . The team struggled in the league though and were relegated two seasons later , due in large part to complacency . However , under the stewardship of manager Joe Mercer Villa returned to the top @-@ flight in 1960 as Second Division Champions . The following season Aston Villa became the first team to win the Football League Cup .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Aston Villa won their sixth FA Cup in 1920 , soon after though the club began a slow decline that led to Villa , at the time one of the most famous and successful clubs in world football , being relegated in 1936 for the first time to the Second Division . This was largely the result of a dismal defensive record : they conceded 110 goals in 42 games , 7 of them coming from Arsenal 's Ted Drake in an infamous 1 – 7 defeat at Villa Park . Like all English clubs , Villa lost seven seasons to the Second World War , and that conflict brought several careers to a premature end . The team was rebuilt under the guidance of former player Alex Massie for the remainder of the 1940s . Aston Villa 's first trophy for 37 years came in the 1956 – 57 season when another former Villa player , Eric Houghton led the club to a then record seventh FA Cup Final win , defeating the ' Busby Babes ' of Manchester United in the final . The team struggled in the league though and were relegated two seasons later , due in large part to complacency . However , under the stewardship of manager Joe Mercer Villa returned to the top @-@ flight in 1960 as Second Division Champions . The following season Aston Villa became the first team to win the Football League Cup . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
Mercer 's forced retirement from the club in 1964 signalled a period of deep turmoil . The most successful club in England was struggling to keep pace with changes in the modern game , with Villa being relegated for the third time , under manager Dick Taylor in 1967 . The following season the fans called for the board to resign as Villa finished 16th in the Second Division . With mounting debts and Villa lying at the bottom of Division Two , the board sacked Tommy Cummings ( the manager brought in to replace Taylor ) , and within weeks the entire board resigned under overwhelming pressure from fans . After much speculation , control of the club was bought by London financier Pat Matthews , who also brought in Doug Ellis as chairman . However , new ownership could not prevent Villa being relegated to the Third Division for the first time at the end of the 1969 – 70 season . However , Villa gradually began to recover under the management of former club captain Vic Crowe . In the 1971 – 72 season they returned to the Second Division as Champions with a record 70 points . In 1974 , Ron Saunders was appointed manager . His brand of no @-@ nonsense man @-@ management proved effective , with the club winning the League Cup the following season and , at the end of season 1974 – 75 , he had taken them back into the First Division and into Europe .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Mercer 's forced retirement from the club in 1964 signalled a period of deep turmoil . The most successful club in England was struggling to keep pace with changes in the modern game , with Villa being relegated for the third time , under manager Dick Taylor in 1967 . The following season the fans called for the board to resign as Villa finished 16th in the Second Division . With mounting debts and Villa lying at the bottom of Division Two , the board sacked Tommy Cummings ( the manager brought in to replace Taylor ) , and within weeks the entire board resigned under overwhelming pressure from fans . After much speculation , control of the club was bought by London financier Pat Matthews , who also brought in Doug Ellis as chairman . However , new ownership could not prevent Villa being relegated to the Third Division for the first time at the end of the 1969 – 70 season . However , Villa gradually began to recover under the management of former club captain Vic Crowe . In the 1971 – 72 season they returned to the Second Division as Champions with a record 70 points . In 1974 , Ron Saunders was appointed manager . His brand of no @-@ nonsense man @-@ management proved effective , with the club winning the League Cup the following season and , at the end of season 1974 – 75 , he had taken them back into the First Division and into Europe . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
Villa were back among the elite as Saunders continued to mould a winning team . This culminated in a seventh top @-@ flight league title in 1980 – 81 . To the surprise of commentators and fans , Saunders quit halfway through the 1981 – 82 season , after falling out with the chairman , with Villa in the quarter final of the European Cup . He was replaced by his softly @-@ spoken assistant manager Tony Barton who guided the club to a 1 – 0 victory over Bayern Munich in the European Cup final in Rotterdam courtesy of a Peter Withe goal . The following season Villa were crowned European Super Cup winners , beating Barcelona in the final . This marked a pinnacle though and Villa 's fortunes declined sharply for most of the 1980s , culminating in relegation in 1987 . This was followed by promotion the following year under Graham Taylor and a runners @-@ up position in the First Division in the 1989 – 90 season .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Villa were back among the elite as Saunders continued to mould a winning team . This culminated in a seventh top @-@ flight league title in 1980 – 81 . To the surprise of commentators and fans , Saunders quit halfway through the 1981 – 82 season , after falling out with the chairman , with Villa in the quarter final of the European Cup . He was replaced by his softly @-@ spoken assistant manager Tony Barton who guided the club to a 1 – 0 victory over Bayern Munich in the European Cup final in Rotterdam courtesy of a Peter Withe goal . The following season Villa were crowned European Super Cup winners , beating Barcelona in the final . This marked a pinnacle though and Villa 's fortunes declined sharply for most of the 1980s , culminating in relegation in 1987 . This was followed by promotion the following year under Graham Taylor and a runners @-@ up position in the First Division in the 1989 – 90 season . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
Villa were one of the founding members of the Premier League in 1992 , and finished runners @-@ up to Manchester United in the inaugural season . For the rest of the Nineties however Villa went through three different managers and their league positions were inconsistent , although they did win two League Cups and regularly achieved UEFA Cup qualification . Villa reached the FA Cup final in 2000 but lost 1 – 0 to Chelsea in the last game to be played at the old Wembley Stadium . Again Villa 's league position continued to fluctuate under several different managers and things came to a head in the summer of 2006 when David O 'Leary left in acrimony . After 23 years as chairman and single biggest shareholder ( approximately 38 % ) , Doug Ellis finally decided to sell his stake in Aston Villa due to ill @-@ health . After much speculation it was announced the club was to be bought by American businessman Randy Lerner , owner of NFL franchise the Cleveland Browns .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Villa were one of the founding members of the Premier League in 1992 , and finished runners @-@ up to Manchester United in the inaugural season . For the rest of the Nineties however Villa went through three different managers and their league positions were inconsistent , although they did win two League Cups and regularly achieved UEFA Cup qualification . Villa reached the FA Cup final in 2000 but lost 1 – 0 to Chelsea in the last game to be played at the old Wembley Stadium . Again Villa 's league position continued to fluctuate under several different managers and things came to a head in the summer of 2006 when David O 'Leary left in acrimony . After 23 years as chairman and single biggest shareholder ( approximately 38 % ) , Doug Ellis finally decided to sell his stake in Aston Villa due to ill @-@ health . After much speculation it was announced the club was to be bought by American businessman Randy Lerner , owner of NFL franchise the Cleveland Browns . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "Villa were one of the founding members of the Premier League in <|start-PI|> 1992 <|end-PI|> <|retrieval|> <|start-KFI|> , and finished runners-up to Manchester United in the inaugural season <|end-KFI|>. For the rest of the Nineties however Villa went through three different managers and their league positions were inconsistent, although they did win two League Cups and regularly achieved UEFA Cup qualification. <|start-PI|> Villa reached the FA Cup final in <|end-PI|> <|retrieval|> <|start-KFI|> 2000 but lost 1-0 to Chelsea in the last game to be played at the old Wembley Stadium <|end-KFI|>. Again Villa's league position continued to fluctuate under several different managers and things came to a head in the summer of 2006 when David O'Leary left in acrimony. <|start-PI|> After 23 years as chairman and single biggest shareholder (approximately 38%), Doug Ellis finally decided to sell his stake in Aston Villa due to <|end-PI|> <|retrieval|> <|start-KFI|> ill-health <|end-KFI|>. <|start-PI|> After much speculation it was announced the club was to be bought by <|end-PI|> <|retrieval|> <|start-KFI|> American businessman Randy Lerner, owner of NFL franchise the Cleveland Browns <|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces Villa's role in the Premier League's founding, while the KFI provides the specific year and their performance. The <|retrieval|> token is inserted to indicate where additional information about the Premier League's formation might be needed.",
"Ideal Knowledge": "A detailed history of the Premier League's formation, including the list of founding members and the final standings of the inaugural 1992-93 season.",
"Key Factual Information": "1992, and finished runners-up to Manchester United in the inaugural season",
"Prefix reformulated": "When did Aston Villa become a founding member of the Premier League?",
"Prefix-intent": "Villa were one of the founding members of the Premier League in",
"Verification": "The KFI directly provides the year of Premier League founding and Villa's performance, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about Villa's FA Cup final appearance, and the KFI provides the specific year, result, and historical context. The <|retrieval|> token is inserted to indicate where additional information about this particular FA Cup final might be needed.",
"Ideal Knowledge": "A comprehensive record of FA Cup finals, including the 2000 final between Aston Villa and Chelsea, with details about the score, venue, and its significance as the last game at the old Wembley Stadium.",
"Key Factual Information": "2000 but lost 1-0 to Chelsea in the last game to be played at the old Wembley Stadium",
"Prefix reformulated": "When did Aston Villa reach the FA Cup final and what was the result?",
"Prefix-intent": "Villa reached the FA Cup final in",
"Verification": "The KFI directly provides the year of the FA Cup final, the result, and additional context about the venue, as prompted by the PI."
},
{
"Explanation": "The PI provides context about Doug Ellis's long-term involvement with Aston Villa and sets up a query about his reason for selling, while the KFI succinctly states the reason. The <|retrieval|> token is inserted to indicate where additional information about Ellis's decision might be needed.",
"Ideal Knowledge": "Biographical information about Doug Ellis, including his tenure as Aston Villa chairman, his shareholding, and the circumstances surrounding his decision to sell the club.",
"Key Factual Information": "ill-health",
"Prefix reformulated": "Why did Doug Ellis decide to sell his stake in Aston Villa after 23 years as chairman?",
"Prefix-intent": "After 23 years as chairman and single biggest shareholder (approximately 38%), Doug Ellis finally decided to sell his stake in Aston Villa due to",
"Verification": "The KFI directly provides the reason for Ellis selling his stake, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the new owner of Aston Villa, and the KFI provides specific information about Randy Lerner and his background. The <|retrieval|> token is inserted to indicate where additional information about Lerner and the club's ownership change might be needed.",
"Ideal Knowledge": "Detailed information about Randy Lerner, including his business background, ownership of the Cleveland Browns, and his acquisition of Aston Villa.",
"Key Factual Information": "American businessman Randy Lerner, owner of NFL franchise the Cleveland Browns",
"Prefix reformulated": "Who bought Aston Villa after Doug Ellis decided to sell?",
"Prefix-intent": "After much speculation it was announced the club was to be bought by",
"Verification": "The KFI directly provides the identity and background of the buyer, as prompted by the PI."
}
]
} |
The arrival of a new owner in Lerner and of manager Martin O 'Neill marked the start of a new period of optimism at Villa Park and sweeping changes occurred throughout the club including a new badge , a new kit sponsor and team changes in the summer of 2007 . The first Cup final of the Lerner era came in 2010 when Villa were beaten 2 – 1 in the League Cup Final . Villa made a second trip to Wembley in that season losing 3 – 0 to Chelsea in the FA Cup semifinal . Just five days before the opening day of the 2010 – 11 season , O 'Neill resigned as manager with immediate effect . The club appointed Gérard Houllier as a replacement in September 2010 , but he stepped down on 1 June 2011 due to ill @-@ health . Houllier was replaced by Birmingham City manager Alex McLeish , despite numerous protests from fans against his appointment ; this was the first time that a manager had moved directly from Birmingham to Villa . McLeish 's contract was terminated at the end of the 2011 – 12 season after Villa finished in 16th place , only just above the relegation zone . On 2 July 2012 , Aston Villa confirmed the appointment of former Norwich City manager Paul Lambert as the replacement for McLeish . On 28 February 2012 , the club announced a financial loss of £ 53 @.@ 9 million . Lerner put the club up for sale on 12 May 2014 , with an estimated value of £ 200 million . With Lerner still on board , in the 2014 – 15 season Aston Villa scored just 12 goals in 25 league games , the lowest in Premier League history , and Lambert was sacked on 11 February 2015 . Tim Sherwood succeeded him , and saved Aston Villa from relegation while also leading them to the 2015 FA Cup Final , but he was sacked in the 2015 – 16 season , as was his successor Rémi Garde . Eric Black took temporary charge of the team , but was not able to prevent Villa from being relegated for the first time since 1987 . On 2 June 2016 , Roberto Di Matteo was announced as the club 's new manager .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n The arrival of a new owner in Lerner and of manager Martin O 'Neill marked the start of a new period of optimism at Villa Park and sweeping changes occurred throughout the club including a new badge , a new kit sponsor and team changes in the summer of 2007 . The first Cup final of the Lerner era came in 2010 when Villa were beaten 2 – 1 in the League Cup Final . Villa made a second trip to Wembley in that season losing 3 – 0 to Chelsea in the FA Cup semifinal . Just five days before the opening day of the 2010 – 11 season , O 'Neill resigned as manager with immediate effect . The club appointed Gérard Houllier as a replacement in September 2010 , but he stepped down on 1 June 2011 due to ill @-@ health . Houllier was replaced by Birmingham City manager Alex McLeish , despite numerous protests from fans against his appointment ; this was the first time that a manager had moved directly from Birmingham to Villa . McLeish 's contract was terminated at the end of the 2011 – 12 season after Villa finished in 16th place , only just above the relegation zone . On 2 July 2012 , Aston Villa confirmed the appointment of former Norwich City manager Paul Lambert as the replacement for McLeish . On 28 February 2012 , the club announced a financial loss of £ 53 @.@ 9 million . Lerner put the club up for sale on 12 May 2014 , with an estimated value of £ 200 million . With Lerner still on board , in the 2014 – 15 season Aston Villa scored just 12 goals in 25 league games , the lowest in Premier League history , and Lambert was sacked on 11 February 2015 . Tim Sherwood succeeded him , and saved Aston Villa from relegation while also leading them to the 2015 FA Cup Final , but he was sacked in the 2015 – 16 season , as was his successor Rémi Garde . Eric Black took temporary charge of the team , but was not able to prevent Villa from being relegated for the first time since 1987 . On 2 June 2016 , Roberto Di Matteo was announced as the club 's new manager . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "The arrival of a new owner in Lerner and of manager Martin O 'Neill marked the start of a new period of optimism at Villa Park and sweeping changes occurred throughout the club including a new badge , a new kit sponsor and team changes in the summer of 2007 . <|start-PI|> The first Cup final of the Lerner era came in <|end-PI|> <|retrieval|> <|start-KFI|> 2010 when Villa were beaten 2 – 1 in the League Cup Final <|end-KFI|> . Villa made a second trip to Wembley in that season losing 3 – 0 to Chelsea in the FA Cup semifinal . <|start-PI|> Just five days before the opening day of the 2010 – 11 season , O 'Neill resigned as manager <|end-PI|> <|retrieval|> <|start-KFI|> with immediate effect <|end-KFI|> . <|start-PI|> The club appointed Gérard Houllier as a replacement in September 2010 , but he stepped down on <|end-PI|> <|retrieval|> <|start-KFI|> 1 June 2011 due to ill @-@ health <|end-KFI|> . Houllier was replaced by Birmingham City manager Alex McLeish , despite numerous protests from fans against his appointment ; this was the first time that a manager had moved directly from Birmingham to Villa . McLeish 's contract was terminated at the end of the 2011 – 12 season after Villa finished in 16th place , only just above the relegation zone . On 2 July 2012 , Aston Villa confirmed the appointment of former Norwich City manager Paul Lambert as the replacement for McLeish . <|start-PI|> On 28 February 2012 , the club announced a financial loss of <|end-PI|> <|retrieval|> <|start-KFI|> £ 53 @.@ 9 million <|end-KFI|> . <|start-PI|> Lerner put the club up for sale on 12 May 2014 , with an estimated value of <|end-PI|> <|retrieval|> <|start-KFI|> £ 200 million <|end-KFI|> . With Lerner still on board , in the 2014 – 15 season Aston Villa scored just 12 goals in 25 league games , the lowest in Premier League history , and Lambert was sacked on 11 February 2015 . Tim Sherwood succeeded him , and saved Aston Villa from relegation while also leading them to the 2015 FA Cup Final , but he was sacked in the 2015 – 16 season , as was his successor Rémi Garde . Eric Black took temporary charge of the team , but was not able to prevent Villa from being relegated for the first time since 1987 . On 2 June 2016 , Roberto Di Matteo was announced as the club 's new manager .",
"none": null,
"results": [
{
"Explanation": "The PI sets up a query about the first Cup final under Lerner, and the KFI provides the specific year and outcome. The <|retrieval|> token is inserted to indicate where additional information about this event might be needed.",
"Ideal Knowledge": "A timeline of Aston Villa's cup performances under Randy Lerner's ownership, specifically highlighting the 2010 League Cup Final and its result.",
"Key Factual Information": "2010 when Villa were beaten 2 – 1 in the League Cup Final",
"Prefix reformulated": "When did the first Cup final of the Lerner era at Aston Villa occur?",
"Prefix-intent": "The first Cup final of the Lerner era came in",
"Verification": "The KFI directly provides the year and result of the first Cup final under Lerner's ownership, as prompted by the PI."
},
{
"Explanation": "The PI provides context about O'Neill's resignation, and the KFI specifies the immediate nature of the resignation. The <|retrieval|> token is inserted to indicate where additional information about this event might be needed.",
"Ideal Knowledge": "Detailed information about Martin O'Neill's tenure as Aston Villa manager, including the circumstances and timing of his resignation just before the 2010-11 season.",
"Key Factual Information": "with immediate effect",
"Prefix reformulated": "When did Martin O'Neill resign as manager of Aston Villa?",
"Prefix-intent": "Just five days before the opening day of the 2010 – 11 season , O 'Neill resigned as manager",
"Verification": "The KFI directly describes the nature of O'Neill's resignation, as prompted by the PI."
},
{
"Explanation": "The PI introduces Houllier's appointment and sets up a query about his departure, while the KFI provides the specific date and reason. The <|retrieval|> token is inserted to indicate where additional information about Houllier's time at Villa might be needed.",
"Ideal Knowledge": "A detailed account of Gérard Houllier's brief tenure as Aston Villa manager, including his appointment date, key events during his management, and the circumstances of his departure due to health issues.",
"Key Factual Information": "1 June 2011 due to ill @-@ health",
"Prefix reformulated": "When did Gérard Houllier step down as manager of Aston Villa?",
"Prefix-intent": "The club appointed Gérard Houllier as a replacement in September 2010 , but he stepped down on",
"Verification": "The KFI directly provides the date and reason for Houllier's departure, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the club's financial announcement, and the KFI provides the specific amount of loss. The <|retrieval|> token is inserted to indicate where additional financial information might be needed.",
"Ideal Knowledge": "Detailed financial reports of Aston Villa for the 2011-2012 fiscal year, including a breakdown of revenue, expenses, and the factors contributing to the significant loss.",
"Key Factual Information": "£ 53 @.@ 9 million",
"Prefix reformulated": "What was the financial loss announced by Aston Villa on 28 February 2012?",
"Prefix-intent": "On 28 February 2012 , the club announced a financial loss of",
"Verification": "The KFI directly provides the amount of financial loss announced by the club, as prompted by the PI."
},
{
"Explanation": "The PI introduces Lerner's decision to sell the club and sets up a query about its value, while the KFI provides the specific estimated amount. The <|retrieval|> token is inserted to indicate where additional information about the club's valuation might be needed.",
"Ideal Knowledge": "Detailed information about Randy Lerner's decision to sell Aston Villa, including market analysis of football club valuations in 2014, factors influencing the £200 million estimate, and comparisons with other club sales in that period.",
"Key Factual Information": "£ 200 million",
"Prefix reformulated": "What was the estimated value of Aston Villa when Randy Lerner put the club up for sale on 12 May 2014?",
"Prefix-intent": "Lerner put the club up for sale on 12 May 2014 , with an estimated value of",
"Verification": "The KFI directly provides the estimated value of the club when it was put up for sale, as prompted by the PI."
}
]
} |
The club colours are a claret shirt with sky blue sleeves , white shorts with claret and blue trim , and sky blue socks with claret and white trim . They were the original wearers of the claret and blue . Villa 's colours at the outset generally comprised plain shirts ( white , grey or a shade of blue ) , with either white or black shorts . For a few years after that ( 1877 – 79 ) the team wore several different kits from all white , blue and black , red and blue to plain green . By 1880 , black jerseys with a red lion embroidered on the chest were introduced by William McGregor . This remained the first choice strip for six years . On Monday , 8 November 1886 , an entry in the club 's official minute book states :
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n The club colours are a claret shirt with sky blue sleeves , white shorts with claret and blue trim , and sky blue socks with claret and white trim . They were the original wearers of the claret and blue . Villa 's colours at the outset generally comprised plain shirts ( white , grey or a shade of blue ) , with either white or black shorts . For a few years after that ( 1877 – 79 ) the team wore several different kits from all white , blue and black , red and blue to plain green . By 1880 , black jerseys with a red lion embroidered on the chest were introduced by William McGregor . This remained the first choice strip for six years . On Monday , 8 November 1886 , an entry in the club 's official minute book states : \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>The club colours are a claret shirt with sky blue sleeves<|end-PI|><|retrieval|><|start-KFI|>, white shorts with claret and blue trim, and sky blue socks with claret and white trim<|end-KFI|>. <|start-PI|>They were the original wearers of<|end-PI|><|retrieval|><|start-KFI|> the claret and blue<|end-KFI|>. <|start-PI|>Villa's colours at the outset generally comprised<|end-PI|><|retrieval|><|start-KFI|> plain shirts (white, grey or a shade of blue), with either white or black shorts<|end-KFI|>. For a few years after that (1877–79) the team wore several different kits from all white, blue and black, red and blue to plain green. <|start-PI|>By 1880, black jerseys with a red lion embroidered on the chest were introduced by<|end-PI|><|retrieval|><|start-KFI|> William McGregor<|end-KFI|>. This remained the first choice strip for six years. On Monday, 8 November 1886, an entry in the club's official minute book states:",
"none": null,
"results": [
{
"Explanation": "The PI introduces the topic of the club colours, and the KFI provides the detailed description. The <|retrieval|> token is inserted to indicate where additional information about the club's colours might be needed.",
"Ideal Knowledge": "A comprehensive description of Aston Villa's official club colours, including the specific shades and design elements of their home kit.",
"Key Factual Information": "claret shirt with sky blue sleeves, white shorts with claret and blue trim, and sky blue socks with claret and white trim",
"Prefix reformulated": "What are the club colours of Aston Villa?",
"Prefix-intent": "The club colours are a claret shirt with sky blue sleeves",
"Verification": "The KFI directly provides the detailed description of the club colours as prompted by the PI."
},
{
"Explanation": "The PI sets up a statement about Aston Villa's original colours, and the KFI provides the specific colour combination. The <|retrieval|> token is inserted to indicate where additional historical information might be needed.",
"Ideal Knowledge": "Historical information about the origin of the claret and blue colours in football, specifically mentioning Aston Villa as the first team to adopt this colour combination.",
"Key Factual Information": "the claret and blue",
"Prefix reformulated": "What colours were Aston Villa the original wearers of?",
"Prefix-intent": "They were the original wearers of",
"Verification": "The KFI directly states the colours that Aston Villa originally wore, as prompted by the PI."
},
{
"Explanation": "The PI introduces the topic of Villa's initial colours, and the KFI provides the specific description. The <|retrieval|> token is inserted to indicate where additional information about the club's early kit history might be needed.",
"Ideal Knowledge": "A chronological list of Aston Villa's kit colours from the club's founding, detailing the various combinations used in the early years.",
"Key Factual Information": "plain shirts (white, grey or a shade of blue), with either white or black shorts",
"Prefix reformulated": "What were Aston Villa's initial colours?",
"Prefix-intent": "Villa's colours at the outset generally comprised",
"Verification": "The KFI directly describes the initial colours worn by Aston Villa, as prompted by the PI."
},
{
"Explanation": "The PI describes a specific jersey design and sets up a query about who introduced it, while the KFI provides the name of the person responsible. The <|retrieval|> token is inserted to indicate where additional information about William McGregor might be needed.",
"Ideal Knowledge": "Biographical information about William McGregor, his role in Aston Villa's history, and his contributions to the club's kit design in 1880.",
"Key Factual Information": "William McGregor",
"Prefix reformulated": "Who introduced black jerseys with a red lion embroidered on the chest for Aston Villa in 1880?",
"Prefix-intent": "By 1880, black jerseys with a red lion embroidered on the chest were introduced by",
"Verification": "The KFI directly provides the name of the person who introduced the described jersey, as prompted by the PI."
}
]
} |
( i ) Proposed and seconded that the colours be chocolate and sky blue shirts and that we order two dozen .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n ( i ) Proposed and seconded that the colours be chocolate and sky blue shirts and that we order two dozen . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
( ii ) Proposed and seconded that Mr McGregor be requested to supply them at the lowest quotation .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n ( ii ) Proposed and seconded that Mr McGregor be requested to supply them at the lowest quotation . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
The chocolate colour later became claret . Nobody is quite sure why claret and blue became the club 's adopted colours . Several other English football teams adopted their colours ; clubs that wear claret and blue include West Ham United and Burnley .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n The chocolate colour later became claret . Nobody is quite sure why claret and blue became the club 's adopted colours . Several other English football teams adopted their colours ; clubs that wear claret and blue include West Ham United and Burnley . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>The chocolate colour later became<|end-PI|> <|retrieval|> <|start-KFI|>claret<|end-KFI|>. <|start-PI|>Nobody is quite sure why claret and blue became<|end-PI|> <|retrieval|> <|start-KFI|>the club's adopted colours<|end-KFI|>. <|start-PI|>Several other English football teams adopted their colours ; clubs that wear claret and blue include<|end-PI|> <|retrieval|> <|start-KFI|>West Ham United and Burnley<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context for a color change, and the KFI provides the specific new color. The <|retrieval|> token is inserted to indicate where additional information about the color change might be needed.",
"Ideal Knowledge": "A detailed history of the club's uniform colors, specifically mentioning the transition from chocolate to claret.",
"Key Factual Information": "claret",
"Prefix reformulated": "What color did the chocolate color of the club's uniform later become?",
"Prefix-intent": "The chocolate colour later became",
"Verification": "The KFI directly provides the color that the chocolate color changed to, as prompted by the PI."
},
{
"Explanation": "The PI introduces uncertainty about the reason for the color choice, and the KFI confirms that these indeed became the club's colors. The <|retrieval|> token is inserted to indicate where additional information about the color adoption might be needed.",
"Ideal Knowledge": "Information on the history and possible theories behind the club's adoption of claret and blue colors, including any cultural or historical significance.",
"Key Factual Information": "the club's adopted colours",
"Prefix reformulated": "Why did claret and blue become the club's adopted colors?",
"Prefix-intent": "Nobody is quite sure why claret and blue became",
"Verification": "The KFI confirms that claret and blue became the club's colors, as implied by the PI."
},
{
"Explanation": "The PI sets up the context for other teams using the same colors, and the KFI provides specific examples. The <|retrieval|> token is inserted to indicate where additional information about other clubs with similar colors might be needed.",
"Ideal Knowledge": "A comprehensive list of English football clubs that use claret and blue as their team colors, including details about when and why they adopted these colors.",
"Key Factual Information": "West Ham United and Burnley",
"Prefix reformulated": "Which other English football clubs wear claret and blue colors?",
"Prefix-intent": "Several other English football teams adopted their colours ; clubs that wear claret and blue include",
"Verification": "The KFI directly provides examples of other clubs wearing claret and blue, as prompted by the PI."
}
]
} |
A new badge was revealed in May 2007 , for the 2007 – 08 season and beyond . The new badge includes a star to represent the European Cup win in 1982 , and has a light blue background behind Villa 's ' lion rampant ' . The traditional motto " Prepared " remains in the badge , and the name Aston Villa has been shortened to AVFC , FC having been omitted from the previous badge . The lion is now unified as opposed to fragmented lions of the past . Randy Lerner petitioned fans to help with the design of the new badge .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n A new badge was revealed in May 2007 , for the 2007 – 08 season and beyond . The new badge includes a star to represent the European Cup win in 1982 , and has a light blue background behind Villa 's ' lion rampant ' . The traditional motto \" Prepared \" remains in the badge , and the name Aston Villa has been shortened to AVFC , FC having been omitted from the previous badge . The lion is now unified as opposed to fragmented lions of the past . Randy Lerner petitioned fans to help with the design of the new badge . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
On 6 April 2016 , the club confirmed that it will be using a new badge from the 2016 – 17 season after consulting fan groups for suggestions . The lion in the new badge will have claws added to it and the word " Prepared " will be removed to increase the size of the lion and club initials in the badge .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n On 6 April 2016 , the club confirmed that it will be using a new badge from the 2016 – 17 season after consulting fan groups for suggestions . The lion in the new badge will have claws added to it and the word \" Prepared \" will be removed to increase the size of the lion and club initials in the badge . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
Aston Villa forwent commercial kit sponsorship for the 2008 – 09 and 2009 – 10 seasons ; instead advertising the charity Acorns Children 's Hospice , the first deal of its kind in Premier league history . The partnership continued until 2010 when a commercial sponsor replaced Acorns , with the hospice becoming the club 's Official Charity Partner . In 2014 – 15 , the Acorns name returned to Aston Villa 's home and away shirts , but only for children 's shirts re @-@ affirming the club 's support for the children 's charity .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Aston Villa forwent commercial kit sponsorship for the 2008 – 09 and 2009 – 10 seasons ; instead advertising the charity Acorns Children 's Hospice , the first deal of its kind in Premier league history . The partnership continued until 2010 when a commercial sponsor replaced Acorns , with the hospice becoming the club 's Official Charity Partner . In 2014 – 15 , the Acorns name returned to Aston Villa 's home and away shirts , but only for children 's shirts re @-@ affirming the club 's support for the children 's charity . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Aston Villa forwent commercial kit sponsorship for the 2008 – 09 and 2009 – 10 seasons ; instead advertising<|end-PI|> <|retrieval|> <|start-KFI|>the charity Acorns Children 's Hospice , the first deal of its kind in Premier league history<|end-KFI|> . <|start-PI|>The partnership continued until 2010 when<|end-PI|> <|retrieval|> <|start-KFI|>a commercial sponsor replaced Acorns , with the hospice becoming the club 's Official Charity Partner<|end-KFI|> . <|start-PI|>In 2014 – 15 , the Acorns name returned to Aston Villa 's home and away shirts , but only for<|end-PI|> <|retrieval|> <|start-KFI|>children 's shirts re @-@ affirming the club 's support for the children 's charity<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of Aston Villa's unique sponsorship decision, while the KFI provides the specific charity and its historical significance. The <|retrieval|> token is inserted to indicate where additional information about this sponsorship deal might be needed.",
"Ideal Knowledge": "Information about Aston Villa's kit sponsorship history, focusing on their charitable partnership with Acorns Children's Hospice and its significance in Premier League history.",
"Key Factual Information": "the charity Acorns Children 's Hospice , the first deal of its kind in Premier league history",
"Prefix reformulated": "What did Aston Villa advertise on their kit instead of commercial sponsorship in the 2008-09 and 2009-10 seasons?",
"Prefix-intent": "Aston Villa forwent commercial kit sponsorship for the 2008 – 09 and 2009 – 10 seasons ; instead advertising",
"Verification": "The KFI directly provides the information about what Aston Villa advertised instead of commercial sponsorship, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the end of the initial partnership, and the KFI provides information about the change in sponsorship and the new role of Acorns. The <|retrieval|> token is inserted to indicate where additional information about this transition might be needed.",
"Ideal Knowledge": "Details about the transition of Aston Villa's sponsorship in 2010, including information about the new commercial sponsor and the continued relationship with Acorns Children's Hospice.",
"Key Factual Information": "a commercial sponsor replaced Acorns , with the hospice becoming the club 's Official Charity Partner",
"Prefix reformulated": "What happened to Aston Villa's partnership with Acorns Children's Hospice in 2010?",
"Prefix-intent": "The partnership continued until 2010 when",
"Verification": "The KFI directly answers what happened to the partnership in 2010, as prompted by the PI."
},
{
"Explanation": "The PI introduces the return of the Acorns name to Aston Villa's shirts, while the KFI specifies the exact nature of this return. The <|retrieval|> token is inserted to indicate where additional information about this specific sponsorship arrangement might be needed.",
"Ideal Knowledge": "Information about Aston Villa's continued support for Acorns Children's Hospice, specifically detailing the 2014-15 season's shirt sponsorship arrangement for children's kits.",
"Key Factual Information": "children 's shirts re @-@ affirming the club 's support for the children 's charity",
"Prefix reformulated": "Where did the Acorns name appear on Aston Villa's shirts in 2014-15?",
"Prefix-intent": "In 2014 – 15 , the Acorns name returned to Aston Villa 's home and away shirts , but only for",
"Verification": "The KFI directly provides information about where the Acorns name appeared on the shirts and why, as prompted by the PI."
}
]
} |
Since 2015 Villa 's shirt sponsors have been Quickbooks . Previous commercial sponsors have been Davenports ( 1982 – 83 ) , Mita ( 1983 – 93 ) , Müller ( 1993 – 95 ) , AST Computer ( 1995 – 98 ) , LDV ( 1998 – 2000 ) , NTL ( 2000 – 02 ) , Rover ( 2002 – 04 ) , DWS Investments ( 2004 – 06 ) , 32Red.com ( 2006 – 08 ) , FxPro ( 2010 – 11 ) , Genting Casinos ( 2011 – 13 ) , Dafabet ( 2013 – 2015 ) , and Intuit QuickBooks ( 2015 – ) . Since 2016 , kit has been manufactured by Under Armour . Previous manufacturers have been Umbro ( 1972 – 81 , 1990 – 93 ) , le Coq Sportif ( 1981 – 83 ) , Henson ( 1983 – 87 ) , Hummel ( 1987 – 90 , 2004 – 07 ) , Asics ( 1993 – 95 ) , Reebok ( 1995 – 2000 ) , Diadora ( 2000 – 04 ) , Nike ( 2007 – 12 ) and Macron ( 2012 @-@ 16 ) .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Since 2015 Villa 's shirt sponsors have been Quickbooks . Previous commercial sponsors have been Davenports ( 1982 – 83 ) , Mita ( 1983 – 93 ) , Müller ( 1993 – 95 ) , AST Computer ( 1995 – 98 ) , LDV ( 1998 – 2000 ) , NTL ( 2000 – 02 ) , Rover ( 2002 – 04 ) , DWS Investments ( 2004 – 06 ) , 32Red.com ( 2006 – 08 ) , FxPro ( 2010 – 11 ) , Genting Casinos ( 2011 – 13 ) , Dafabet ( 2013 – 2015 ) , and Intuit QuickBooks ( 2015 – ) . Since 2016 , kit has been manufactured by Under Armour . Previous manufacturers have been Umbro ( 1972 – 81 , 1990 – 93 ) , le Coq Sportif ( 1981 – 83 ) , Henson ( 1983 – 87 ) , Hummel ( 1987 – 90 , 2004 – 07 ) , Asics ( 1993 – 95 ) , Reebok ( 1995 – 2000 ) , Diadora ( 2000 – 04 ) , Nike ( 2007 – 12 ) and Macron ( 2012 @-@ 16 ) . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
Aston Villa 's current home venue is Villa Park , which is a UEFA 5 @-@ star rated stadium , having previously played at Aston Park ( 1874 – 1876 ) and Wellington Road ( 1876 – 1897 ) . Villa Park is the largest football stadium in the English Midlands , and the eighth largest stadium in England . It has hosted 16 England internationals at senior level , the first in 1899 , and the most recent in 2005 . Thus , it was the first English ground to stage international football in three different centuries . Villa Park is the most used stadium in FA Cup semi @-@ final history , having hosted 55 semi @-@ finals . The club have planning permission to extend the North Stand ; this will involve the ' filling in ' of the corners to either side of the North Stand . If completed , the capacity of Villa Park will be increased to approximately 51 @,@ 000 .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Aston Villa 's current home venue is Villa Park , which is a UEFA 5 @-@ star rated stadium , having previously played at Aston Park ( 1874 – 1876 ) and Wellington Road ( 1876 – 1897 ) . Villa Park is the largest football stadium in the English Midlands , and the eighth largest stadium in England . It has hosted 16 England internationals at senior level , the first in 1899 , and the most recent in 2005 . Thus , it was the first English ground to stage international football in three different centuries . Villa Park is the most used stadium in FA Cup semi @-@ final history , having hosted 55 semi @-@ finals . The club have planning permission to extend the North Stand ; this will involve the ' filling in ' of the corners to either side of the North Stand . If completed , the capacity of Villa Park will be increased to approximately 51 @,@ 000 . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "Aston Villa's <|start-PI|>current home venue is<|end-PI|> <|retrieval|> <|start-KFI|>Villa Park, which is a UEFA 5-star rated stadium<|end-KFI|>, having previously played at Aston Park (1874 – 1876) and Wellington Road (1876 – 1897). <|start-PI|>Villa Park is the largest football stadium in<|end-PI|> <|retrieval|> <|start-KFI|>the English Midlands, and the eighth largest stadium in England<|end-KFI|>. It has hosted 16 England internationals at senior level, the first in 1899, and the most recent in 2005. Thus, it was the first English ground to stage international football in three different centuries. <|start-PI|>Villa Park is the most used stadium in FA Cup semi-final history, having hosted<|end-PI|> <|retrieval|> <|start-KFI|>55 semi-finals<|end-KFI|>. The club have planning permission to extend the North Stand; this will involve the 'filling in' of the corners to either side of the North Stand. <|start-PI|>If completed, the capacity of Villa Park will be increased to approximately<|end-PI|> <|retrieval|> <|start-KFI|>51,000<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces a query about Aston Villa's current home venue, while the KFI provides the specific name and rating. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "Information about Aston Villa's home stadiums throughout history, focusing on Villa Park, including its UEFA rating and significance in English football.",
"Key Factual Information": "Villa Park, which is a UEFA 5-star rated stadium",
"Prefix reformulated": "What is Aston Villa's current home venue?",
"Prefix-intent": "Aston Villa's current home venue is",
"Verification": "The KFI directly provides the name and rating of the stadium requested in the PI."
},
{
"Explanation": "The PI sets up a query about Villa Park's size ranking, and the KFI provides the specific regional and national rankings. The <|retrieval|> token is inserted to indicate where additional information about stadium sizes might be needed.",
"Ideal Knowledge": "A comprehensive list of football stadiums in England, ranked by size, with specific focus on the Midlands region and Villa Park's position in these rankings.",
"Key Factual Information": "the English Midlands, and the eighth largest stadium in England",
"Prefix reformulated": "In which region is Villa Park the largest football stadium?",
"Prefix-intent": "Villa Park is the largest football stadium in",
"Verification": "The KFI directly answers the question about Villa Park's size ranking in the region and country, as prompted by the PI."
},
{
"Explanation": "The PI introduces Villa Park's significance in FA Cup history and sets up a query about the number of semi-finals hosted, while the KFI provides the specific number. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "Detailed statistics on FA Cup semi-final venues throughout history, highlighting Villa Park's prominence and the specific number of semi-finals it has hosted.",
"Key Factual Information": "55 semi-finals",
"Prefix reformulated": "How many FA Cup semi-finals has Villa Park hosted?",
"Prefix-intent": "Villa Park is the most used stadium in FA Cup semi-final history, having hosted",
"Verification": "The KFI directly provides the number of semi-finals hosted at Villa Park, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the future capacity of Villa Park after planned extensions, and the KFI provides the specific expected capacity. The <|retrieval|> token is inserted to indicate where additional information about the stadium expansion plans might be needed.",
"Ideal Knowledge": "Detailed plans for Villa Park's extension, including current capacity, proposed changes, and the resulting increase in spectator capacity.",
"Key Factual Information": "51,000",
"Prefix reformulated": "What will be the approximate capacity of Villa Park after the planned extension?",
"Prefix-intent": "If completed, the capacity of Villa Park will be increased to approximately",
"Verification": "The KFI directly provides the expected capacity after the planned extension, as prompted by the PI."
}
]
} |
The current training ground is located at Bodymoor Heath near Kingsbury in north Warwickshire , the site for which was purchased by former chairman Doug Ellis in the early 1970s from a local farmer . Although Bodymoor Heath was state @-@ of @-@ the @-@ art in the 1970s , by the late 1990s the facilities had started to look dated . In November 2005 , Ellis and Aston Villa plc announced a state of the art GB £ 13 million redevelopment of Bodymoor in two phases . However , work on Bodymoor was suspended by Ellis due to financial problems , and was left in an unfinished state until new owner Randy Lerner made it one of his priorities to make the site one of the best in world football . The new training ground was officially unveiled on 6 May 2007 , by then manager Martin O 'Neill , then team captain Gareth Barry and 1982 European Cup winning team captain Dennis Mortimer , with the Aston Villa squad moving in for the 2007 – 08 season .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n The current training ground is located at Bodymoor Heath near Kingsbury in north Warwickshire , the site for which was purchased by former chairman Doug Ellis in the early 1970s from a local farmer . Although Bodymoor Heath was state @-@ of @-@ the @-@ art in the 1970s , by the late 1990s the facilities had started to look dated . In November 2005 , Ellis and Aston Villa plc announced a state of the art GB £ 13 million redevelopment of Bodymoor in two phases . However , work on Bodymoor was suspended by Ellis due to financial problems , and was left in an unfinished state until new owner Randy Lerner made it one of his priorities to make the site one of the best in world football . The new training ground was officially unveiled on 6 May 2007 , by then manager Martin O 'Neill , then team captain Gareth Barry and 1982 European Cup winning team captain Dennis Mortimer , with the Aston Villa squad moving in for the 2007 – 08 season . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "The current training ground is <|start-PI|>located at<|end-PI|> <|retrieval|> <|start-KFI|>Bodymoor Heath near Kingsbury in north Warwickshire<|end-KFI|>, <|start-PI|>the site for which was purchased by former chairman Doug Ellis in<|end-PI|> <|retrieval|> <|start-KFI|>the early 1970s from a local farmer<|end-KFI|>. Although Bodymoor Heath was state @-@ of @-@ the @-@ art in the 1970s, by the late 1990s the facilities had started to look dated. <|start-PI|>In November 2005, Ellis and Aston Villa plc announced a state of the art<|end-PI|> <|retrieval|> <|start-KFI|>GB £ 13 million redevelopment of Bodymoor in two phases<|end-KFI|>. However, work on Bodymoor was suspended by Ellis due to financial problems, and was left in an unfinished state until new owner Randy Lerner made it one of his priorities to make the site one of the best in world football. <|start-PI|>The new training ground was officially unveiled on<|end-PI|> <|retrieval|> <|start-KFI|>6 May 2007, by then manager Martin O'Neill, then team captain Gareth Barry and 1982 European Cup winning team captain Dennis Mortimer<|end-KFI|>, with the Aston Villa squad moving in for the 2007 – 08 season.",
"none": null,
"results": [
{
"Explanation": "The PI introduces the topic of the training ground's location, while the KFI provides the specific place. The <|retrieval|> token is inserted to indicate where additional information about the location might be needed.",
"Ideal Knowledge": "A detailed description of Aston Villa's training facilities, including their location, history, and surrounding area.",
"Key Factual Information": "Bodymoor Heath near Kingsbury in north Warwickshire",
"Prefix reformulated": "Where is Aston Villa's current training ground located?",
"Prefix-intent": "The current training ground is located at",
"Verification": "The KFI directly provides the location information requested in the PI."
},
{
"Explanation": "The PI sets up the context of the site's purchase, and the KFI provides the specific time and seller. The <|retrieval|> token is inserted to indicate where additional historical information might be needed.",
"Ideal Knowledge": "A timeline of Aston Villa's property acquisitions, focusing on Doug Ellis's tenure as chairman and the purchase of Bodymoor Heath.",
"Key Factual Information": "the early 1970s from a local farmer",
"Prefix reformulated": "When did former Aston Villa chairman Doug Ellis purchase the Bodymoor Heath site?",
"Prefix-intent": "The site for which was purchased by former chairman Doug Ellis in",
"Verification": "The KFI directly provides the time period and source of the purchase as prompted by the PI."
},
{
"Explanation": "The PI introduces a specific announcement, and the KFI provides the details of the redevelopment plan. The <|retrieval|> token is inserted to indicate where additional information about the project might be needed.",
"Ideal Knowledge": "Detailed plans and budget for the 2005 Bodymoor Heath redevelopment project, including the proposed phases and expected improvements.",
"Key Factual Information": "GB £ 13 million redevelopment of Bodymoor in two phases",
"Prefix reformulated": "What did Ellis and Aston Villa plc announce in November 2005?",
"Prefix-intent": "In November 2005, Ellis and Aston Villa plc announced a state of the art",
"Verification": "The KFI directly provides the details of the announcement mentioned in the PI."
},
{
"Explanation": "The PI sets up the context of the training ground's unveiling, and the KFI provides the specific date and important figures present. The <|retrieval|> token is inserted to indicate where additional information about the event might be needed.",
"Ideal Knowledge": "A detailed account of the unveiling ceremony of Aston Villa's new training ground, including the date, attendees, and their significance to the club.",
"Key Factual Information": "6 May 2007, by then manager Martin O'Neill, then team captain Gareth Barry and 1982 European Cup winning team captain Dennis Mortimer",
"Prefix reformulated": "When was Aston Villa's new training ground officially unveiled?",
"Prefix-intent": "The new training ground was officially unveiled on",
"Verification": "The KFI directly provides the date and key figures present at the unveiling, as prompted by the PI."
}
]
} |
It was announced on 6 August 2014 , that Villa Park would appear in the FIFA video game from FIFA 15 onwards , with all other Premier League stadiums also fully licensed from this game onwards .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n It was announced on 6 August 2014 , that Villa Park would appear in the FIFA video game from FIFA 15 onwards , with all other Premier League stadiums also fully licensed from this game onwards . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>It was announced on 6 August 2014 , that Villa Park would appear in the FIFA video game from<|end-PI|> <|retrieval|> <|start-KFI|>FIFA 15 onwards<|end-KFI|> , <|start-PI|>with all other Premier League stadiums also fully licensed from<|end-PI|> <|retrieval|> <|start-KFI|>this game onwards<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of the announcement and when it was made, while the KFI provides the specific FIFA game version from which Villa Park would be included. The <|retrieval|> token is inserted to indicate where additional information about the FIFA game series might be needed.",
"Ideal Knowledge": "A timeline of FIFA video game releases, highlighting when specific stadiums were added to the game, with particular focus on the inclusion of Villa Park and other Premier League stadiums in FIFA 15.",
"Key Factual Information": "FIFA 15 onwards",
"Prefix reformulated": "When was it announced that Villa Park would appear in the FIFA video game series?",
"Prefix-intent": "It was announced on 6 August 2014 , that Villa Park would appear in the FIFA video game from",
"Verification": "The KFI directly provides the specific FIFA game version mentioned in the PI."
},
{
"Explanation": "The PI introduces the topic of Premier League stadium licensing in the FIFA games, and the KFI confirms that all stadiums were licensed from the same game version mentioned earlier (FIFA 15). The <|retrieval|> token is inserted to indicate where additional information about FIFA game licensing might be needed.",
"Ideal Knowledge": "Detailed information about the licensing agreements between FIFA video game series and the Premier League, specifically highlighting the inclusion of all Premier League stadiums from FIFA 15 onwards.",
"Key Factual Information": "this game onwards",
"Prefix reformulated": "From which FIFA game version were all Premier League stadiums fully licensed?",
"Prefix-intent": "with all other Premier League stadiums also fully licensed from",
"Verification": "The KFI directly answers the question posed in the PI, referring to the previously mentioned FIFA 15."
}
]
} |
The first shares in the club were issued towards the end of the 19th century as a result of legislation that was intended to codify the growing numbers of professional teams and players in the Association Football leagues . FA teams were required to distribute shares to investors as a way of facilitating trading among the teams without implicating the FA itself . This trading continued for much of the 20th century until Ellis started buying up many of the shares in the 1960s . He was chairman and substantial shareholder of " Aston Villa F.C. " from 1968 to 1975 and the majority shareholder from 1982 to 2006 . The club were floated on the London Stock Exchange ( LSE ) in 1996 , and the share price fluctuated in the ten years after the flotation . In 2006 it was announced that several consortia and individuals were considering bids for Aston Villa .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n The first shares in the club were issued towards the end of the 19th century as a result of legislation that was intended to codify the growing numbers of professional teams and players in the Association Football leagues . FA teams were required to distribute shares to investors as a way of facilitating trading among the teams without implicating the FA itself . This trading continued for much of the 20th century until Ellis started buying up many of the shares in the 1960s . He was chairman and substantial shareholder of \" Aston Villa F.C. \" from 1968 to 1975 and the majority shareholder from 1982 to 2006 . The club were floated on the London Stock Exchange ( LSE ) in 1996 , and the share price fluctuated in the ten years after the flotation . In 2006 it was announced that several consortia and individuals were considering bids for Aston Villa . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "The first shares in the club were issued towards the end of the 19th century as a result of <|start-PI|> legislation that was intended to codify <|end-PI|> <|retrieval|> <|start-KFI|> the growing numbers of professional teams and players in the Association Football leagues <|end-KFI|>. <|start-PI|> FA teams were required to distribute shares to investors as a way of <|end-PI|> <|retrieval|> <|start-KFI|> facilitating trading among the teams without implicating the FA itself <|end-KFI|>. This trading continued for much of the 20th century until <|start-PI|> Ellis started buying up many of the shares in <|end-PI|> <|retrieval|> <|start-KFI|> the 1960s <|end-KFI|>. <|start-PI|> He was chairman and substantial shareholder of \"Aston Villa F.C.\" from <|end-PI|> <|retrieval|> <|start-KFI|> 1968 to 1975 and the majority shareholder from 1982 to 2006 <|end-KFI|>. <|start-PI|> The club were floated on the London Stock Exchange ( LSE ) in <|end-PI|> <|retrieval|> <|start-KFI|> 1996 <|end-KFI|>, and the share price fluctuated in the ten years after the flotation. In 2006 it was announced that several consortia and individuals were considering bids for Aston Villa.",
"none": null,
"results": [
{
"Explanation": "The PI introduces the issuance of shares and sets up a query about its cause, while the KFI provides the specific reason. The <|retrieval|> token is inserted to indicate where additional information about the legislation might be needed.",
"Ideal Knowledge": "Information about the late 19th-century legislation affecting professional football clubs in England, specifically detailing the requirements for issuing shares and the reasons behind this legislation.",
"Key Factual Information": "legislation that was intended to codify the growing numbers of professional teams and players in the Association Football leagues",
"Prefix reformulated": "Why were the first shares in Aston Villa Football Club issued towards the end of the 19th century?",
"Prefix-intent": "The first shares in the club were issued towards the end of the 19th century as a result of",
"Verification": "The KFI directly explains the reason for issuing shares, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the purpose of share distribution, and the KFI provides the specific reason. The <|retrieval|> token is inserted to indicate where additional information about FA regulations might be needed.",
"Ideal Knowledge": "Detailed explanation of the Football Association's regulations regarding share distribution among teams, including the rationale behind keeping the FA separate from team trading activities.",
"Key Factual Information": "facilitating trading among the teams without implicating the FA itself",
"Prefix reformulated": "Why were Football Association teams required to distribute shares to investors?",
"Prefix-intent": "FA teams were required to distribute shares to investors as a way of",
"Verification": "The KFI directly provides the purpose of share distribution, as prompted by the PI."
},
{
"Explanation": "The PI introduces Ellis's share acquisition and sets up a query about when it occurred, while the KFI provides the specific time period. The <|retrieval|> token is inserted to indicate where additional information about Ellis's involvement might be needed.",
"Ideal Knowledge": "A timeline of Aston Villa's ownership changes, focusing on Doug Ellis's involvement and share acquisitions in the 1960s.",
"Key Factual Information": "the 1960s",
"Prefix reformulated": "When did Ellis start buying up many of the shares in Aston Villa Football Club?",
"Prefix-intent": "Ellis started buying up many of the shares in",
"Verification": "The KFI directly provides the time period when Ellis started buying shares, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about Ellis's roles and tenure, and the KFI provides the specific time periods. The <|retrieval|> token is inserted to indicate where additional information about Ellis's career at Aston Villa might be needed.",
"Ideal Knowledge": "A detailed chronology of Doug Ellis's involvement with Aston Villa, including his various positions and shareholding status from 1968 to 2006.",
"Key Factual Information": "1968 to 1975 and the majority shareholder from 1982 to 2006",
"Prefix reformulated": "What period was Ellis chairman and substantial shareholder of Aston Villa Football Club?",
"Prefix-intent": "He was chairman and substantial shareholder of \"Aston Villa F.C.\" from",
"Verification": "The KFI directly provides the time periods of Ellis's roles, as prompted by the PI."
},
{
"Explanation": "The PI introduces the club's flotation on the LSE and sets up a query about when it occurred, while the KFI provides the specific year. The <|retrieval|> token is inserted to indicate where additional information about the flotation might be needed.",
"Ideal Knowledge": "Information about Aston Villa's initial public offering in 1996, including details about the flotation process and its impact on the club's financial structure.",
"Key Factual Information": "1996",
"Prefix reformulated": "When was Aston Villa Football Club floated on the London Stock Exchange?",
"Prefix-intent": "The club were floated on the London Stock Exchange ( LSE ) in",
"Verification": "The KFI directly provides the year of the club's flotation on the LSE, as prompted by the PI."
}
]
} |
On 14 August 2006 , it was confirmed that Randy Lerner , then owner of the National Football League 's Cleveland Browns , had reached an agreement of £ 62 @.@ 6 million with Aston Villa for a takeover of the club . A statement released on 25 August to the LSE announced that Lerner had secured 59 @.@ 69 % of Villa shares , making him the majority shareholder . He also appointed himself Chairman of the club . In Ellis 's last year in charge Villa lost £ 8.2m before tax , compared with a £ 3m profit the previous year , and income had fallen from £ 51.6m to £ 49m . Lerner took full control on 18 September , as he had 89 @.@ 69 % of the shares . On 19 September 2006 , Ellis and his board resigned to be replaced with a new board headed by Lerner . Lerner installed Charles Krulak as a non @-@ executive director and Ellis was awarded the honorary position of Chairman Emeritus .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n On 14 August 2006 , it was confirmed that Randy Lerner , then owner of the National Football League 's Cleveland Browns , had reached an agreement of £ 62 @.@ 6 million with Aston Villa for a takeover of the club . A statement released on 25 August to the LSE announced that Lerner had secured 59 @.@ 69 % of Villa shares , making him the majority shareholder . He also appointed himself Chairman of the club . In Ellis 's last year in charge Villa lost £ 8.2m before tax , compared with a £ 3m profit the previous year , and income had fallen from £ 51.6m to £ 49m . Lerner took full control on 18 September , as he had 89 @.@ 69 % of the shares . On 19 September 2006 , Ellis and his board resigned to be replaced with a new board headed by Lerner . Lerner installed Charles Krulak as a non @-@ executive director and Ellis was awarded the honorary position of Chairman Emeritus . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "On 14 August 2006, <|start-PI|>it was confirmed that Randy Lerner, then owner of the National Football League's Cleveland Browns, had reached an agreement of<|end-PI|> <|retrieval|> <|start-KFI|>£62.6 million with Aston Villa for a takeover of the club<|end-KFI|>. <|start-PI|>A statement released on 25 August to the LSE announced that Lerner had secured<|end-PI|> <|retrieval|> <|start-KFI|>59.69% of Villa shares, making him the majority shareholder<|end-KFI|>. He also appointed himself Chairman of the club. <|start-PI|>In Ellis's last year in charge Villa lost<|end-PI|> <|retrieval|> <|start-KFI|>£8.2m before tax, compared with a £3m profit the previous year, and income had fallen from £51.6m to £49m<|end-KFI|>. <|start-PI|>Lerner took full control on 18 September, as he had<|end-PI|> <|retrieval|> <|start-KFI|>89.69% of the shares<|end-KFI|>. On 19 September 2006, Ellis and his board resigned to be replaced with a new board headed by Lerner. Lerner installed Charles Krulak as a non-executive director and Ellis was awarded the honorary position of Chairman Emeritus.",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of Randy Lerner's agreement, and the KFI provides the specific monetary value. The <|retrieval|> token is inserted to indicate where additional information about the takeover deal might be needed.",
"Ideal Knowledge": "Detailed information about Randy Lerner's acquisition of Aston Villa, including the exact amount paid, date of the agreement, and Lerner's background as the owner of the Cleveland Browns.",
"Key Factual Information": "£62.6 million with Aston Villa for a takeover of the club",
"Prefix reformulated": "What was the value of the agreement Randy Lerner reached for Aston Villa's takeover?",
"Prefix-intent": "On 14 August 2006, it was confirmed that Randy Lerner, then owner of the National Football League's Cleveland Browns, had reached an agreement of",
"Verification": "The KFI directly provides the monetary value of the agreement mentioned in the PI."
},
{
"Explanation": "The PI introduces the LSE statement about Lerner's share acquisition, and the KFI provides the specific percentage and its implication. The <|retrieval|> token is inserted to indicate where additional information about the shareholding might be needed.",
"Ideal Knowledge": "Detailed breakdown of Aston Villa's shareholding structure after Randy Lerner's acquisition, including the exact percentage he acquired and its significance in terms of control over the club.",
"Key Factual Information": "59.69% of Villa shares, making him the majority shareholder",
"Prefix reformulated": "What percentage of Aston Villa shares did Randy Lerner secure according to the LSE statement?",
"Prefix-intent": "A statement released on 25 August to the LSE announced that Lerner had secured",
"Verification": "The KFI directly provides the percentage of shares Lerner secured, as mentioned in the PI."
},
{
"Explanation": "The PI sets up a query about Villa's financial performance under Ellis, and the KFI provides specific financial figures. The <|retrieval|> token is inserted to indicate where additional financial information might be needed.",
"Ideal Knowledge": "Detailed financial statements of Aston Villa for the years before and during the ownership transition, including profit/loss figures, income changes, and reasons for the financial downturn.",
"Key Factual Information": "£8.2m before tax, compared with a £3m profit the previous year, and income had fallen from £51.6m to £49m",
"Prefix reformulated": "How much did Aston Villa lose in Ellis's last year in charge?",
"Prefix-intent": "In Ellis's last year in charge Villa lost",
"Verification": "The KFI directly provides the financial loss and compares it to the previous year's profit, as prompted by the PI."
},
{
"Explanation": "The PI introduces Lerner taking full control, and the KFI provides the specific percentage of shares he held at that time. The <|retrieval|> token is inserted to indicate where additional information about the takeover process might be needed.",
"Ideal Knowledge": "Detailed timeline of Randy Lerner's acquisition of Aston Villa shares, including the initial purchase, subsequent share acquisitions, and the final percentage that gave him full control.",
"Key Factual Information": "89.69% of the shares",
"Prefix reformulated": "What percentage of shares did Randy Lerner have when he took full control of Aston Villa?",
"Prefix-intent": "Lerner took full control on 18 September, as he had",
"Verification": "The KFI directly provides the percentage of shares Lerner had when taking full control, as prompted by the PI."
}
]
} |
On 18 May 2016 , Randy Lerner agreed the sale of Aston Villa to Recon Group , owned by Chinese businessman Xia Jiantong . The sale was completed on 14 June 2016 for a reported £ 76 million after being approved by the Football League , with the club becoming part of Recon Group 's Sport , Leisure and Tourism division . Recon Group were selected to take over Aston Villa following a selection process by the club .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n On 18 May 2016 , Randy Lerner agreed the sale of Aston Villa to Recon Group , owned by Chinese businessman Xia Jiantong . The sale was completed on 14 June 2016 for a reported £ 76 million after being approved by the Football League , with the club becoming part of Recon Group 's Sport , Leisure and Tourism division . Recon Group were selected to take over Aston Villa following a selection process by the club . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>On 18 May 2016, Randy Lerner agreed the sale of Aston Villa to<|end-PI|> <|retrieval|> <|start-KFI|>Recon Group, owned by Chinese businessman Xia Jiantong<|end-KFI|>. <|start-PI|>The sale was completed on 14 June 2016 for<|end-PI|> <|retrieval|> <|start-KFI|>a reported £76 million after being approved by the Football League<|end-KFI|>, <|start-PI|>with the club becoming part of Recon Group's<|end-PI|> <|retrieval|> <|start-KFI|>Sport, Leisure and Tourism division<|end-KFI|>. Recon Group were selected to take over Aston Villa following a selection process by the club.",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of the sale agreement, while the KFI provides the specific details of the buyer. The <|retrieval|> token is inserted to indicate where additional information about the buyer might be needed.",
"Ideal Knowledge": "Information about the ownership changes of Aston Villa Football Club, specifically detailing the sale agreement between Randy Lerner and Recon Group in May 2016, including the ownership of Recon Group by Xia Jiantong.",
"Key Factual Information": "Recon Group, owned by Chinese businessman Xia Jiantong",
"Prefix reformulated": "Who did Randy Lerner agree to sell Aston Villa to on 18 May 2016?",
"Prefix-intent": "On 18 May 2016, Randy Lerner agreed the sale of Aston Villa to",
"Verification": "The KFI directly provides the information about who Randy Lerner agreed to sell Aston Villa to, as prompted by the PI."
},
{
"Explanation": "The PI introduces the completion of the sale, while the KFI provides the specific financial details and regulatory approval. The <|retrieval|> token is inserted to indicate where additional information about the sale terms might be needed.",
"Ideal Knowledge": "Detailed information about the Aston Villa sale transaction, including the exact date of completion, the reported sale price, and the approval process by the Football League.",
"Key Factual Information": "a reported £76 million after being approved by the Football League",
"Prefix reformulated": "What was the reported sale price of Aston Villa on 14 June 2016?",
"Prefix-intent": "The sale was completed on 14 June 2016 for",
"Verification": "The KFI directly provides the reported sale price and additional context about the approval, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of Aston Villa's new position within Recon Group, while the KFI provides the specific division. The <|retrieval|> token is inserted to indicate where additional information about Recon Group's structure might be needed.",
"Ideal Knowledge": "Information about Recon Group's corporate structure, specifically detailing its Sport, Leisure and Tourism division and how Aston Villa fits into this organizational framework.",
"Key Factual Information": "Sport, Leisure and Tourism division",
"Prefix reformulated": "What division of Recon Group did Aston Villa become part of?",
"Prefix-intent": "with the club becoming part of Recon Group's",
"Verification": "The KFI directly specifies the division of Recon Group that Aston Villa joined, as prompted by the PI."
}
]
} |
Aston Villa have a unique relationship with the Acorns Children 's Hospice charity that is groundbreaking in English football . In a first for the Premier League , Aston Villa donated the front of the shirt on their kit , usually reserved for high @-@ paying sponsorships , to Acorns Hospice so that the charity would gain significant additional visibility and greater fund raising capabilities . Outside of the shirt sponsorship the club have paid for hospice care for the charity as well as regularly providing player visits to hospice locations .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Aston Villa have a unique relationship with the Acorns Children 's Hospice charity that is groundbreaking in English football . In a first for the Premier League , Aston Villa donated the front of the shirt on their kit , usually reserved for high @-@ paying sponsorships , to Acorns Hospice so that the charity would gain significant additional visibility and greater fund raising capabilities . Outside of the shirt sponsorship the club have paid for hospice care for the charity as well as regularly providing player visits to hospice locations . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
In September 2010 , Aston Villa launched an initiative at Villa Park called Villa Midlands Food ( VMF ) where the club will spend two years training students with Aston Villa Hospitality and Events in association with Birmingham City Council . The club will open a restaurant in the Trinity Road Stand staffed with 12 students recruited from within a ten @-@ mile ( 16 km ) radius of Villa Park with the majority of the food served in the restaurant sourced locally .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n In September 2010 , Aston Villa launched an initiative at Villa Park called Villa Midlands Food ( VMF ) where the club will spend two years training students with Aston Villa Hospitality and Events in association with Birmingham City Council . The club will open a restaurant in the Trinity Road Stand staffed with 12 students recruited from within a ten @-@ mile ( 16 km ) radius of Villa Park with the majority of the food served in the restaurant sourced locally . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
Aston Villa have a large fanbase and draw support from all over the Midlands and beyond , with supporters ' clubs all across the world . Former Villa chief executive Richard Fitzgerald has stated that the ethnicity of the supporters is currently 98 % white . When Randy Lerner 's regime took over at Villa Park , they aimed to improve their support from ethnic minorities . A number of organisations have been set up to support the local community including Aston Pride . A Villa in the Community programme has also been set up to encourage support among young people in the region . The new owners have also initiated several surveys aimed at gaining the opinions of Villa fans and to involve them in the decision making process . Meetings also occur every three months where supporters are invited by ballot and are invited to ask questions to the Board . In 2011 , the club supported a supporter @-@ based initiative for an official anthem to boost the atmosphere at Villa Park . The song " The Bells Are Ringing " is to be played before games .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Aston Villa have a large fanbase and draw support from all over the Midlands and beyond , with supporters ' clubs all across the world . Former Villa chief executive Richard Fitzgerald has stated that the ethnicity of the supporters is currently 98 % white . When Randy Lerner 's regime took over at Villa Park , they aimed to improve their support from ethnic minorities . A number of organisations have been set up to support the local community including Aston Pride . A Villa in the Community programme has also been set up to encourage support among young people in the region . The new owners have also initiated several surveys aimed at gaining the opinions of Villa fans and to involve them in the decision making process . Meetings also occur every three months where supporters are invited by ballot and are invited to ask questions to the Board . In 2011 , the club supported a supporter @-@ based initiative for an official anthem to boost the atmosphere at Villa Park . The song \" The Bells Are Ringing \" is to be played before games . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
Like many English football clubs Aston Villa have had several hooligan firms associated with them : Villa Youth , Steamers , Villa Hardcore and the C @-@ Crew , the last mentioned being very active during the 1970s and 1980s . As can be seen across the whole of English football , the hooligan groups have now been marginalised . In 2004 , several Villa firms were involved in a fight with QPR fans outside Villa Park in which a steward died . The main groupings of supporters can now be found in a number of domestic and international supporters ' clubs . This includes the Official Aston Villa Supporters Club which also has many smaller regional and international sections . There were several independent supporters clubs during the reign of Doug Ellis but most of these disbanded after his retirement . The supporter group My Old Man Said formed to stand up for Villa supporter 's rights , as a direct result of Villa supporters ' protest against the club 's appointment of Alex McLeish . The club 's supporters also publish fanzines such as Heroes and Villains and The Holy Trinity .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Like many English football clubs Aston Villa have had several hooligan firms associated with them : Villa Youth , Steamers , Villa Hardcore and the C @-@ Crew , the last mentioned being very active during the 1970s and 1980s . As can be seen across the whole of English football , the hooligan groups have now been marginalised . In 2004 , several Villa firms were involved in a fight with QPR fans outside Villa Park in which a steward died . The main groupings of supporters can now be found in a number of domestic and international supporters ' clubs . This includes the Official Aston Villa Supporters Club which also has many smaller regional and international sections . There were several independent supporters clubs during the reign of Doug Ellis but most of these disbanded after his retirement . The supporter group My Old Man Said formed to stand up for Villa supporter 's rights , as a direct result of Villa supporters ' protest against the club 's appointment of Alex McLeish . The club 's supporters also publish fanzines such as Heroes and Villains and The Holy Trinity . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
Aston Villa 's arch @-@ rivals are Birmingham City , with games between the two clubs known as the Second City Derby . Historically though , West Bromwich Albion have arguably been Villa 's greatest rivals , a view highlighted in a fan survey , conducted in 2003 . The two teams contested three FA Cup finals in the late 19th century . Villa also enjoy less heated local rivalries with Wolverhampton Wanderers and Coventry City . Through the relegation of West Brom and Birmingham City , to the Football League Championship , in the 2005 – 06 season , at the start of 2006 – 07 Premiership season , Villa were the only Midlands club in that League . The nearest opposing team Villa faced during that season was Sheffield United , who played 62 miles ( 100 km ) away in South Yorkshire . For the 2010 – 11 season , West Bromwich Albion were promoted and joined Aston Villa , Wolverhampton Wanderers , and Birmingham City in the Premier League . This marked the first time that the " West Midlands ' Big Four " clubs have been in the Premier League at the same time , and the first time together in the top flight since the 1983 – 84 season . Birmingham were relegated at the end of the 2010 – 11 season , ending this period .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Aston Villa 's arch @-@ rivals are Birmingham City , with games between the two clubs known as the Second City Derby . Historically though , West Bromwich Albion have arguably been Villa 's greatest rivals , a view highlighted in a fan survey , conducted in 2003 . The two teams contested three FA Cup finals in the late 19th century . Villa also enjoy less heated local rivalries with Wolverhampton Wanderers and Coventry City . Through the relegation of West Brom and Birmingham City , to the Football League Championship , in the 2005 – 06 season , at the start of 2006 – 07 Premiership season , Villa were the only Midlands club in that League . The nearest opposing team Villa faced during that season was Sheffield United , who played 62 miles ( 100 km ) away in South Yorkshire . For the 2010 – 11 season , West Bromwich Albion were promoted and joined Aston Villa , Wolverhampton Wanderers , and Birmingham City in the Premier League . This marked the first time that the \" West Midlands ' Big Four \" clubs have been in the Premier League at the same time , and the first time together in the top flight since the 1983 – 84 season . Birmingham were relegated at the end of the 2010 – 11 season , ending this period . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
As of the end of the 2014 – 15 season , Aston Villa have spent 104 seasons in the top tier of English football ; the only club to have spent longer in the top flight are Everton , with 112 seasons , making Aston Villa versus Everton the most @-@ played fixture in English top @-@ flight football . Aston Villa were in an elite group of seven clubs that has played in every Premier League season , the other six being Tottenham Hotspur , Chelsea , Everton , Liverpool , Manchester United and Arsenal since its establishment in 1992 – 93 until they were relegated in 2016 . They are seventh in the All @-@ time FA Premier League table , and have the fifth highest total of major honours won by an English club with 21 wins .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n As of the end of the 2014 – 15 season , Aston Villa have spent 104 seasons in the top tier of English football ; the only club to have spent longer in the top flight are Everton , with 112 seasons , making Aston Villa versus Everton the most @-@ played fixture in English top @-@ flight football . Aston Villa were in an elite group of seven clubs that has played in every Premier League season , the other six being Tottenham Hotspur , Chelsea , Everton , Liverpool , Manchester United and Arsenal since its establishment in 1992 – 93 until they were relegated in 2016 . They are seventh in the All @-@ time FA Premier League table , and have the fifth highest total of major honours won by an English club with 21 wins . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>As of the end of the 2014 – 15 season , Aston Villa have spent<|end-PI|> <|retrieval|> <|start-KFI|>104 seasons in the top tier of English football<|end-KFI|> ; <|start-PI|>the only club to have spent longer in the top flight are Everton , with<|end-PI|> <|retrieval|> <|start-KFI|>112 seasons<|end-KFI|> , making Aston Villa versus Everton the most @-@ played fixture in English top @-@ flight football . <|start-PI|>Aston Villa were in an elite group of seven clubs that has played in every Premier League season , the other six being<|end-PI|> <|retrieval|> <|start-KFI|>Tottenham Hotspur , Chelsea , Everton , Liverpool , Manchester United and Arsenal<|end-KFI|> since its establishment in 1992 – 93 until they were relegated in 2016 . <|start-PI|>They are seventh in the All @-@ time FA Premier League table , and have the fifth highest total of major honours won by an English club with<|end-PI|> <|retrieval|> <|start-KFI|>21 wins<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context for Aston Villa's history in top-tier English football, while the KFI provides the specific number of seasons. The <|retrieval|> token is inserted to indicate where additional information about Aston Villa's tenure might be needed.",
"Ideal Knowledge": "A comprehensive list of English football clubs and their tenure in the top tier, with specific focus on Aston Villa's 104 seasons up to the 2014-15 season.",
"Key Factual Information": "104 seasons in the top tier of English football",
"Prefix reformulated": "How many seasons has Aston Villa spent in the top tier of English football?",
"Prefix-intent": "As of the end of the 2014 – 15 season , Aston Villa have spent",
"Verification": "The KFI directly provides the number of seasons Aston Villa has spent in the top tier, as prompted by the PI."
},
{
"Explanation": "The PI introduces Everton as the club with the longest tenure in the top flight, and the KFI provides the specific number of seasons. The <|retrieval|> token is inserted to indicate where additional information about Everton's record might be needed.",
"Ideal Knowledge": "A detailed record of Everton's history in the top flight of English football, highlighting their 112 seasons and comparing it to other long-standing clubs.",
"Key Factual Information": "112 seasons",
"Prefix reformulated": "How many seasons has Everton spent in the top flight of English football?",
"Prefix-intent": "the only club to have spent longer in the top flight are Everton , with",
"Verification": "The KFI directly provides the number of seasons Everton has spent in the top flight, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of Aston Villa being part of an elite group, and the KFI lists the other members of this group. The <|retrieval|> token is inserted to indicate where additional information about these clubs' Premier League participation might be needed.",
"Ideal Knowledge": "A comprehensive list of the seven clubs that have participated in every Premier League season since its establishment in 1992-93, with details about their consistent performance and status.",
"Key Factual Information": "Tottenham Hotspur , Chelsea , Everton , Liverpool , Manchester United and Arsenal",
"Prefix reformulated": "Which clubs have played in every Premier League season along with Aston Villa?",
"Prefix-intent": "Aston Villa were in an elite group of seven clubs that has played in every Premier League season , the other six being",
"Verification": "The KFI directly lists the other six clubs that have played in every Premier League season, as prompted by the PI."
},
{
"Explanation": "The PI sets up Aston Villa's position in the all-time Premier League table and introduces their major honours, while the KFI provides the specific number of wins. The <|retrieval|> token is inserted to indicate where additional information about Aston Villa's honours might be needed.",
"Ideal Knowledge": "A detailed list of all major honours won by English football clubs, with specific focus on Aston Villa's 21 wins and their ranking in comparison to other clubs.",
"Key Factual Information": "21 wins",
"Prefix reformulated": "How many major honours has Aston Villa won as an English club?",
"Prefix-intent": "They are seventh in the All @-@ time FA Premier League table , and have the fifth highest total of major honours won by an English club with",
"Verification": "The KFI directly provides the number of major honours won by Aston Villa, as prompted by the PI."
}
]
} |
Aston Villa currently hold the record number of league goals scored by any team in the English top flight ; 128 goals were scored in the 1930 – 31 season , one more than Arsenal who won the league that season for the very first time , with Villa runners @-@ up . Villa legend Archie Hunter became the first player to score in every round of the FA Cup in Villa 's victorious 1887 campaign . Villa 's longest unbeaten home run in the FA Cup spanned 13 years and 19 games , from 1888 to 1901 .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Aston Villa currently hold the record number of league goals scored by any team in the English top flight ; 128 goals were scored in the 1930 – 31 season , one more than Arsenal who won the league that season for the very first time , with Villa runners @-@ up . Villa legend Archie Hunter became the first player to score in every round of the FA Cup in Villa 's victorious 1887 campaign . Villa 's longest unbeaten home run in the FA Cup spanned 13 years and 19 games , from 1888 to 1901 . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Aston Villa currently hold the record number of league goals scored by any team in the English top flight ;<|end-PI|> <|retrieval|> <|start-KFI|>128 goals were scored in the 1930 – 31 season<|end-KFI|> , one more than Arsenal who won the league that season for the very first time , with Villa runners @-@ up . <|start-PI|>Villa legend Archie Hunter became the first player to score in every round of the FA Cup in<|end-PI|> <|retrieval|> <|start-KFI|>Villa 's victorious 1887 campaign<|end-KFI|> . <|start-PI|>Villa 's longest unbeaten home run in the FA Cup spanned<|end-PI|> <|retrieval|> <|start-KFI|>13 years and 19 games , from 1888 to 1901<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces Aston Villa's goal-scoring record, and the KFI provides the specific number and season. The <|retrieval|> token is inserted to indicate where additional information about this record might be needed.",
"Ideal Knowledge": "A comprehensive list of goal-scoring records in English top-flight football, highlighting Aston Villa's achievement in the 1930-31 season with specific goal tallies and comparisons to other teams.",
"Key Factual Information": "128 goals were scored in the 1930 – 31 season",
"Prefix reformulated": "What is the record number of league goals scored by Aston Villa in the English top flight?",
"Prefix-intent": "Aston Villa currently hold the record number of league goals scored by any team in the English top flight ;",
"Verification": "The KFI directly provides the number of goals and the season, answering the question posed in the PI."
},
{
"Explanation": "The PI introduces Archie Hunter's achievement, and the KFI specifies the year and outcome of the campaign. The <|retrieval|> token is inserted to indicate where additional information about this historic achievement might be needed.",
"Ideal Knowledge": "Detailed information about Archie Hunter's career, focusing on his 1887 FA Cup performance, including match-by-match goal statistics and the significance of scoring in every round of the competition.",
"Key Factual Information": "Villa 's victorious 1887 campaign",
"Prefix reformulated": "In which year did Aston Villa legend Archie Hunter become the first player to score in every round of the FA Cup?",
"Prefix-intent": "Villa legend Archie Hunter became the first player to score in every round of the FA Cup in",
"Verification": "The KFI directly provides the year of the campaign in which Archie Hunter achieved this feat, as prompted by the PI."
},
{
"Explanation": "The PI introduces Villa's longest unbeaten home run in the FA Cup, and the KFI provides the specific duration and timeframe. The <|retrieval|> token is inserted to indicate where additional information about this record might be needed.",
"Ideal Knowledge": "A comprehensive record of Aston Villa's FA Cup home performances, detailing the unbeaten streak from 1888 to 1901, including match results, opponents faced, and any notable events during this period.",
"Key Factual Information": "13 years and 19 games , from 1888 to 1901",
"Prefix reformulated": "How long was Aston Villa's longest unbeaten home run in the FA Cup?",
"Prefix-intent": "Villa 's longest unbeaten home run in the FA Cup spanned",
"Verification": "The KFI directly provides the duration, number of games, and specific years of the unbeaten run, answering the question posed in the PI."
}
]
} |
Aston Villa are one of five English teams to have won the European Cup . They did so on 26 May 1982 in Rotterdam , beating Bayern Munich 1 – 0 thanks to Peter Withe 's goal .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Aston Villa are one of five English teams to have won the European Cup . They did so on 26 May 1982 in Rotterdam , beating Bayern Munich 1 – 0 thanks to Peter Withe 's goal . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Aston Villa are one of five English teams to have won<|end-PI|> <|retrieval|> <|start-KFI|>the European Cup<|end-KFI|>. <|start-PI|>They did so on 26 May 1982 in<|end-PI|> <|retrieval|> <|start-KFI|>Rotterdam<|end-KFI|>, <|start-PI|>beating Bayern Munich 1 – 0 thanks to<|end-PI|> <|retrieval|> <|start-KFI|>Peter Withe 's goal<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces Aston Villa and sets up a query about their major European achievement, while the KFI provides the specific trophy. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "A list of English football clubs that have won the European Cup/Champions League, including Aston Villa and the years they won it.",
"Key Factual Information": "the European Cup",
"Prefix reformulated": "What major European trophy has Aston Villa won?",
"Prefix-intent": "Aston Villa are one of five English teams to have won",
"Verification": "The KFI directly provides the trophy information requested in the PI."
},
{
"Explanation": "The PI sets up the context of Aston Villa's European Cup win with the date, and the KFI provides the specific location. The <|retrieval|> token is inserted to indicate where additional information about the match details might be needed.",
"Ideal Knowledge": "Details of the 1982 European Cup final, including the date, location, teams involved, and the result.",
"Key Factual Information": "Rotterdam",
"Prefix reformulated": "Where did Aston Villa win the European Cup on 26 May 1982?",
"Prefix-intent": "They did so on 26 May 1982 in",
"Verification": "The KFI directly provides the location information requested in the PI."
},
{
"Explanation": "The PI sets up the context of the match result and implies a query about the goal scorer, while the KFI provides the specific player who scored. The <|retrieval|> token is inserted to indicate where additional information about the goal and the player might be needed.",
"Ideal Knowledge": "Detailed match report of the 1982 European Cup final, focusing on Peter Withe's winning goal for Aston Villa against Bayern Munich.",
"Key Factual Information": "Peter Withe 's goal",
"Prefix reformulated": "Who scored the winning goal for Aston Villa in the 1982 European Cup final against Bayern Munich?",
"Prefix-intent": "beating Bayern Munich 1 – 0 thanks to",
"Verification": "The KFI directly provides the goal scorer information implied by the PI."
}
]
} |
Aston Villa have won European and domestic league honours . The club 's last major honour was in 1996 when they won the League Cup .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Aston Villa have won European and domestic league honours . The club 's last major honour was in 1996 when they won the League Cup . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
First Division 7 : 1893 – 94 , 1895 – 96 , 1896 – 97 , 1898 – 99 , 1899 – 1900 , 1909 – 10 , 1980 – 81
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n First Division 7 : 1893 – 94 , 1895 – 96 , 1896 – 97 , 1898 – 99 , 1899 – 1900 , 1909 – 10 , 1980 – 81 \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
Note : Flags indicate national team as defined under FIFA eligibility rules . Players may hold more than one non @-@ FIFA nationality .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Note : Flags indicate national team as defined under FIFA eligibility rules . Players may hold more than one non @-@ FIFA nationality . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
Note : Flags indicate national team as defined under FIFA eligibility rules . Players may hold more than one non @-@ FIFA nationality .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Note : Flags indicate national team as defined under FIFA eligibility rules . Players may hold more than one non @-@ FIFA nationality . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
Note : Flags indicate national team as defined under FIFA eligibility rules . Players may hold more than one non @-@ FIFA nationality .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Note : Flags indicate national team as defined under FIFA eligibility rules . Players may hold more than one non @-@ FIFA nationality . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Note : Flags indicate national team as defined under<|end-PI|> <|retrieval|> <|start-KFI|>FIFA eligibility rules<|end-KFI|>. <|start-PI|>Players may hold more than one<|end-PI|> <|retrieval|> <|start-KFI|>non @-@ FIFA nationality<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces the concept of flags indicating national teams, while the KFI specifies the governing rules. The <|retrieval|> token is inserted to indicate where additional information about FIFA eligibility rules might be needed.",
"Ideal Knowledge": "A detailed explanation of FIFA eligibility rules, including how they relate to national team representation and the use of flags in international football.",
"Key Factual Information": "FIFA eligibility rules",
"Prefix reformulated": "What do flags indicate for national teams?",
"Prefix-intent": "Note : Flags indicate national team as defined under",
"Verification": "The KFI directly provides the specific rules that define national team flags, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about players holding multiple nationalities, and the KFI specifies the type of nationality. The <|retrieval|> token is inserted to indicate where additional information about non-FIFA nationalities might be needed.",
"Ideal Knowledge": "An explanation of the concept of non-FIFA nationalities in football, including examples of players with multiple nationalities and how this affects their eligibility for national teams.",
"Key Factual Information": "non @-@ FIFA nationality",
"Prefix reformulated": "What type of nationality can players hold more than one of?",
"Prefix-intent": "Players may hold more than one",
"Verification": "The KFI directly answers what type of multiple nationalities players can hold, as prompted by the PI."
}
]
} |
There have been many players who can be called notable throughout Aston Villa 's history . These can be classified and recorded in several forms . The Halls of Fame and PFA Players of the Year are noted below . As of 2014 , Aston Villa , jointly with Tottenham Hotspur , hold the record for providing the most England internationals with 73 . Aston Villa have had several players who were one @-@ club men . In 1998 , to celebrate the 100th season of League football , The Football League released a list entitled the Football League 100 Legends that consisted of " 100 legendary football players . " There were seven players included on the list who had formerly played for Villa : Danny Blanchflower , Trevor Ford , Archie Hunter , Sam Hardy , Paul McGrath , Peter Schmeichel and Clem Stephenson .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n There have been many players who can be called notable throughout Aston Villa 's history . These can be classified and recorded in several forms . The Halls of Fame and PFA Players of the Year are noted below . As of 2014 , Aston Villa , jointly with Tottenham Hotspur , hold the record for providing the most England internationals with 73 . Aston Villa have had several players who were one @-@ club men . In 1998 , to celebrate the 100th season of League football , The Football League released a list entitled the Football League 100 Legends that consisted of \" 100 legendary football players . \" There were seven players included on the list who had formerly played for Villa : Danny Blanchflower , Trevor Ford , Archie Hunter , Sam Hardy , Paul McGrath , Peter Schmeichel and Clem Stephenson . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
Three Aston Villa players have won the PFA Players ' Player of the Year award . At the end of every English football season , the members of the Professional Footballers ' Association ( PFA ) vote on which of its members has played the best football in the previous year . In 1977 Andy Gray won the award . In 1990 it was awarded to David Platt , whilst Paul McGrath won it in 1993 . The PFA Young Player of the Year , which is awarded to players under the age of 23 , has been awarded to four players from Aston Villa : Andy Gray in 1977 ; Gary Shaw in 1981 ; Ashley Young in 2009 and James Milner in 2010 . The National Football Museum in Preston , Lancashire administers the English Football Hall of Fame which currently contains two Villa teams , two Villa players and one manager . The 1890s team and 1982 team were inducted into the Hall of Fame in July 2009 . Joe Mercer was inducted into the Hall of Fame at the same time for his career as a manager including his time at Aston Villa . The only two Villa players in the Hall of Fame are Danny Blanchflower and Peter Schmeichel .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Three Aston Villa players have won the PFA Players ' Player of the Year award . At the end of every English football season , the members of the Professional Footballers ' Association ( PFA ) vote on which of its members has played the best football in the previous year . In 1977 Andy Gray won the award . In 1990 it was awarded to David Platt , whilst Paul McGrath won it in 1993 . The PFA Young Player of the Year , which is awarded to players under the age of 23 , has been awarded to four players from Aston Villa : Andy Gray in 1977 ; Gary Shaw in 1981 ; Ashley Young in 2009 and James Milner in 2010 . The National Football Museum in Preston , Lancashire administers the English Football Hall of Fame which currently contains two Villa teams , two Villa players and one manager . The 1890s team and 1982 team were inducted into the Hall of Fame in July 2009 . Joe Mercer was inducted into the Hall of Fame at the same time for his career as a manager including his time at Aston Villa . The only two Villa players in the Hall of Fame are Danny Blanchflower and Peter Schmeichel . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Three Aston Villa players have won the PFA Players ' Player of the Year award<|end-PI|> <|retrieval|> <|start-KFI|>In 1977 Andy Gray won the award . In 1990 it was awarded to David Platt , whilst Paul McGrath won it in 1993<|end-KFI|> . At the end of every English football season , the members of the Professional Footballers ' Association ( PFA ) vote on which of its members has played the best football in the previous year . <|start-PI|>The PFA Young Player of the Year , which is awarded to players under the age of 23 , has been awarded to four players from Aston Villa<|end-PI|> <|retrieval|> <|start-KFI|>Andy Gray in 1977 ; Gary Shaw in 1981 ; Ashley Young in 2009 and James Milner in 2010<|end-KFI|> . <|start-PI|>The National Football Museum in Preston , Lancashire administers the English Football Hall of Fame which currently contains two Villa teams , two Villa players and one manager<|end-PI|> <|retrieval|> <|start-KFI|>The 1890s team and 1982 team were inducted into the Hall of Fame in July 2009 . Joe Mercer was inducted into the Hall of Fame at the same time for his career as a manager including his time at Aston Villa . The only two Villa players in the Hall of Fame are Danny Blanchflower and Peter Schmeichel<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the fact that three Aston Villa players won the award, and the KFI provides the specific names and years. The <|retrieval|> token is inserted to indicate where additional information about the award winners might be needed.",
"Ideal Knowledge": "A comprehensive list of PFA Players' Player of the Year award winners, specifically highlighting Aston Villa players and the years they won.",
"Key Factual Information": "In 1977 Andy Gray won the award . In 1990 it was awarded to David Platt , whilst Paul McGrath won it in 1993",
"Prefix reformulated": "Which Aston Villa players have won the PFA Players' Player of the Year award?",
"Prefix-intent": "Three Aston Villa players have won the PFA Players ' Player of the Year award",
"Verification": "The KFI directly provides the names and years of Aston Villa players who won the award, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context for the PFA Young Player of the Year award and mentions that four Aston Villa players have won it. The KFI provides the specific names and years. The <|retrieval|> token is inserted to indicate where additional information about these winners might be needed.",
"Ideal Knowledge": "A detailed list of PFA Young Player of the Year award winners, focusing on Aston Villa players and including the years they won.",
"Key Factual Information": "Andy Gray in 1977 ; Gary Shaw in 1981 ; Ashley Young in 2009 and James Milner in 2010",
"Prefix reformulated": "Which Aston Villa players have won the PFA Young Player of the Year award?",
"Prefix-intent": "The PFA Young Player of the Year , which is awarded to players under the age of 23 , has been awarded to four players from Aston Villa",
"Verification": "The KFI directly lists the Aston Villa players who won the PFA Young Player of the Year award and their respective years, as prompted by the PI."
},
{
"Explanation": "The PI introduces the English Football Hall of Fame and mentions the number of Villa-related inductees. The KFI provides specific details about these inductees. The <|retrieval|> token is inserted to indicate where additional information about the Hall of Fame and its inductees might be needed.",
"Ideal Knowledge": "A comprehensive list of Aston Villa-related inductees in the English Football Hall of Fame, including teams, players, and managers, with their induction dates and reasons for inclusion.",
"Key Factual Information": "The 1890s team and 1982 team were inducted into the Hall of Fame in July 2009 . Joe Mercer was inducted into the Hall of Fame at the same time for his career as a manager including his time at Aston Villa . The only two Villa players in the Hall of Fame are Danny Blanchflower and Peter Schmeichel",
"Prefix reformulated": "What Aston Villa teams, players, and managers are in the English Football Hall of Fame?",
"Prefix-intent": "The National Football Museum in Preston , Lancashire administers the English Football Hall of Fame which currently contains two Villa teams , two Villa players and one manager",
"Verification": "The KFI directly provides the information about Aston Villa teams, players, and managers in the Hall of Fame, as prompted by the PI."
}
]
} |
In 2006 , Aston Villa announced the creation of an " Aston Villa Hall of Fame . " This was voted for by fans and the inaugural induction saw 12 former players , managers and directors named . In May 2013 it was announced that former Villa and Bulgaria captain , Stiliyan Petrov , would be the 13th addition to the Hall of Fame .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n In 2006 , Aston Villa announced the creation of an \" Aston Villa Hall of Fame . \" This was voted for by fans and the inaugural induction saw 12 former players , managers and directors named . In May 2013 it was announced that former Villa and Bulgaria captain , Stiliyan Petrov , would be the 13th addition to the Hall of Fame . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>In 2006, Aston Villa announced the creation of an<|end-PI|> <|retrieval|> <|start-KFI|>\"Aston Villa Hall of Fame.\"<|end-KFI|> <|start-PI|>This was voted for by fans and the inaugural induction saw<|end-PI|> <|retrieval|> <|start-KFI|>12 former players, managers and directors named.<|end-KFI|> <|start-PI|>In May 2013 it was announced that former Villa and Bulgaria captain,<|end-PI|> <|retrieval|> <|start-KFI|>Stiliyan Petrov, would be the 13th addition to the Hall of Fame.<|end-KFI|>",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context for the announcement, while the KFI provides the specific information about what was created. The <|retrieval|> token is inserted to indicate where additional information about the Hall of Fame might be needed.",
"Ideal Knowledge": "Detailed information about the Aston Villa Hall of Fame, including its establishment date, purpose, and significance to the club's history.",
"Key Factual Information": "\"Aston Villa Hall of Fame.\"",
"Prefix reformulated": "What did Aston Villa announce the creation of in 2006?",
"Prefix-intent": "In 2006, Aston Villa announced the creation of an",
"Verification": "The KFI directly provides the information about what Aston Villa created, as prompted by the PI."
},
{
"Explanation": "The PI introduces the voting process and sets up a query about the inaugural induction, while the KFI provides the specific number and categories of inductees. The <|retrieval|> token is inserted to indicate where additional information about the inductees might be needed.",
"Ideal Knowledge": "A list of the 12 inaugural inductees into the Aston Villa Hall of Fame, including their names, roles, and contributions to the club.",
"Key Factual Information": "12 former players, managers and directors named.",
"Prefix reformulated": "How many former players, managers and directors were named in the inaugural induction of the Aston Villa Hall of Fame?",
"Prefix-intent": "This was voted for by fans and the inaugural induction saw",
"Verification": "The KFI directly provides the number and types of people inducted, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context for a new addition to the Hall of Fame, while the KFI provides the specific name and details. The <|retrieval|> token is inserted to indicate where additional information about Stiliyan Petrov might be needed.",
"Ideal Knowledge": "Biographical information about Stiliyan Petrov, including his career at Aston Villa, his role as captain for both Villa and Bulgaria, and his achievements that led to his induction into the Hall of Fame.",
"Key Factual Information": "Stiliyan Petrov, would be the 13th addition to the Hall of Fame.",
"Prefix reformulated": "Who was announced as the 13th addition to the Aston Villa Hall of Fame in May 2013?",
"Prefix-intent": "In May 2013 it was announced that former Villa and Bulgaria captain,",
"Verification": "The KFI directly provides the name of the person added to the Hall of Fame, as prompted by the PI."
}
]
} |
The following managers have all won at least one trophy when in charge or have been notable for Villa in the context of the League , for example Jozef Vengloš who holds a League record .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n The following managers have all won at least one trophy when in charge or have been notable for Villa in the context of the League , for example Jozef Vengloš who holds a League record . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
A number of television programmes have included references to Aston Villa over the past few decades . In the sitcom Porridge , the character Lennie Godber is a Villa supporter . When filming began on Dad 's Army , Villa fan Ian Lavender was allowed to choose Frank Pike 's scarf from an array in the BBC wardrobe ; he chose a claret and blue one — Aston Villa 's colours . The character Nessa in the BBC sitcom Gavin & Stacey was revealed as an Aston Villa fan in an episode screened in December 2009 . In the BBC series " Yes Minister " / " Yes Prime Minister " , the Minister Jim Hacker 's local team was Aston Villa , as was the Member of Parliament for Birmingham East .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n A number of television programmes have included references to Aston Villa over the past few decades . In the sitcom Porridge , the character Lennie Godber is a Villa supporter . When filming began on Dad 's Army , Villa fan Ian Lavender was allowed to choose Frank Pike 's scarf from an array in the BBC wardrobe ; he chose a claret and blue one — Aston Villa 's colours . The character Nessa in the BBC sitcom Gavin & Stacey was revealed as an Aston Villa fan in an episode screened in December 2009 . In the BBC series \" Yes Minister \" / \" Yes Prime Minister \" , the Minister Jim Hacker 's local team was Aston Villa , as was the Member of Parliament for Birmingham East . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "A number of television programmes have included references to Aston Villa over the past few decades. <|start-PI|>In the sitcom Porridge, the character Lennie Godber is<|end-PI|> <|retrieval|> <|start-KFI|>a Villa supporter<|end-KFI|>. <|start-PI|>When filming began on Dad's Army, Villa fan Ian Lavender was allowed to choose Frank Pike's scarf from an array in the BBC wardrobe; he chose<|end-PI|> <|retrieval|> <|start-KFI|>a claret and blue one — Aston Villa's colours<|end-KFI|>. <|start-PI|>The character Nessa in the BBC sitcom Gavin & Stacey was revealed as<|end-PI|> <|retrieval|> <|start-KFI|>an Aston Villa fan in an episode screened in December 2009<|end-KFI|>. <|start-PI|>In the BBC series \"Yes Minister\" / \"Yes Prime Minister\", the Minister Jim Hacker's local team was<|end-PI|> <|retrieval|> <|start-KFI|>Aston Villa, as was the Member of Parliament for Birmingham East<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces the character and sets up a query about his football allegiance, while the KFI provides the specific team. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "Information about the character Lennie Godber from the sitcom Porridge, including his background and his support for Aston Villa football club.",
"Key Factual Information": "a Villa supporter",
"Prefix reformulated": "What team does Lennie Godber support in the sitcom Porridge?",
"Prefix-intent": "In the sitcom Porridge, the character Lennie Godber is",
"Verification": "The KFI directly states Lennie Godber's team affiliation, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of Ian Lavender choosing a scarf for his character, and the KFI provides the specific colors and their significance. The <|retrieval|> token is inserted to indicate where additional information about the scarf choice and team colors might be needed.",
"Ideal Knowledge": "Details about the costume choices in Dad's Army, specifically Ian Lavender's selection of Frank Pike's scarf, and information about Aston Villa's team colors.",
"Key Factual Information": "a claret and blue one — Aston Villa's colours",
"Prefix reformulated": "What color scarf did Ian Lavender choose for Frank Pike in Dad's Army?",
"Prefix-intent": "When filming began on Dad's Army, Villa fan Ian Lavender was allowed to choose Frank Pike's scarf from an array in the BBC wardrobe; he chose",
"Verification": "The KFI directly provides the color of the scarf chosen and its connection to Aston Villa, as prompted by the PI."
},
{
"Explanation": "The PI introduces the character and sets up a query about her football allegiance, while the KFI provides the specific team and when it was revealed. The <|retrieval|> token is inserted to indicate where additional information about the character and the episode might be needed.",
"Ideal Knowledge": "Information about the character Nessa from Gavin & Stacey, including her revealed support for Aston Villa and the specific episode where this information was disclosed.",
"Key Factual Information": "an Aston Villa fan in an episode screened in December 2009",
"Prefix reformulated": "What team does Nessa support in the BBC sitcom Gavin & Stacey?",
"Prefix-intent": "The character Nessa in the BBC sitcom Gavin & Stacey was revealed as",
"Verification": "The KFI directly states Nessa's team affiliation and when it was revealed, as prompted by the PI."
},
{
"Explanation": "The PI introduces the character and sets up a query about his local team, while the KFI provides the specific team and additional context. The <|retrieval|> token is inserted to indicate where additional information about the character and the show's setting might be needed.",
"Ideal Knowledge": "Information about the character Jim Hacker from \"Yes Minister\" / \"Yes Prime Minister\", including his local football team affiliation and its connection to his constituency.",
"Key Factual Information": "Aston Villa, as was the Member of Parliament for Birmingham East",
"Prefix reformulated": "What is Minister Jim Hacker's local team in \"Yes Minister\" / \"Yes Prime Minister\"?",
"Prefix-intent": "In the BBC series \"Yes Minister\" / \"Yes Prime Minister\", the Minister Jim Hacker's local team was",
"Verification": "The KFI directly states Jim Hacker's local team and provides additional context, as prompted by the PI."
}
]
} |
Aston Villa have also featured on several occasions in prose . Stanley Woolley , a character in Derek Robinson 's Booker shortlisted novel Goshawk Squadron is an Aston Villa fan and names a pre @-@ war starting eleven Villa side . Together with The Oval , Villa Park is referenced by the poet Philip Larkin in his poem about the First World War , MCMXIV . Aston Villa are also mentioned in Harold Pinter 's play The Dumb Waiter .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Aston Villa have also featured on several occasions in prose . Stanley Woolley , a character in Derek Robinson 's Booker shortlisted novel Goshawk Squadron is an Aston Villa fan and names a pre @-@ war starting eleven Villa side . Together with The Oval , Villa Park is referenced by the poet Philip Larkin in his poem about the First World War , MCMXIV . Aston Villa are also mentioned in Harold Pinter 's play The Dumb Waiter . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "Aston Villa have also featured on several occasions in prose. <|start-PI|>Stanley Woolley, a character in Derek Robinson's Booker shortlisted novel Goshawk Squadron is<|end-PI|> <|retrieval|> <|start-KFI|>an Aston Villa fan and names a pre-war starting eleven Villa side<|end-KFI|>. <|start-PI|>Together with The Oval, Villa Park is referenced by the poet Philip Larkin in his poem about<|end-PI|> <|retrieval|> <|start-KFI|>the First World War, MCMXIV<|end-KFI|>. <|start-PI|>Aston Villa are also mentioned in Harold Pinter's play<|end-PI|> <|retrieval|> <|start-KFI|>The Dumb Waiter<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces Stanley Woolley and sets up a query about his characteristics, while the KFI provides specific information about his football affiliation. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "Details about the character Stanley Woolley from the novel Goshawk Squadron, including his support for Aston Villa and his knowledge of the team's pre-war lineup.",
"Key Factual Information": "an Aston Villa fan and names a pre-war starting eleven Villa side",
"Prefix reformulated": "What is Stanley Woolley's affiliation in Derek Robinson's novel Goshawk Squadron?",
"Prefix-intent": "Stanley Woolley, a character in Derek Robinson's Booker shortlisted novel Goshawk Squadron is",
"Verification": "The KFI directly provides information about Stanley Woolley's connection to Aston Villa, as prompted by the PI."
},
{
"Explanation": "The PI mentions Villa Park being referenced in a poem by Philip Larkin, and the KFI provides the specific context and title of the poem. The <|retrieval|> token is inserted to indicate where additional information about the poem might be needed.",
"Ideal Knowledge": "Information about Philip Larkin's poem MCMXIV, including its subject matter (the First World War) and its references to specific locations like Villa Park and The Oval.",
"Key Factual Information": "the First World War, MCMXIV",
"Prefix reformulated": "In which poem does Philip Larkin reference Villa Park and The Oval?",
"Prefix-intent": "Together with The Oval, Villa Park is referenced by the poet Philip Larkin in his poem about",
"Verification": "The KFI directly provides the subject and title of the poem that references Villa Park, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about a play by Harold Pinter that mentions Aston Villa, and the KFI provides the specific title of the play. The <|retrieval|> token is inserted to indicate where additional information about the play might be needed.",
"Ideal Knowledge": "Information about Harold Pinter's play The Dumb Waiter, including its plot summary and the context in which Aston Villa is mentioned.",
"Key Factual Information": "The Dumb Waiter",
"Prefix reformulated": "In which play by Harold Pinter is Aston Villa mentioned?",
"Prefix-intent": "Aston Villa are also mentioned in Harold Pinter's play",
"Verification": "The KFI directly provides the title of the play that mentions Aston Villa, as prompted by the PI."
}
]
} |
Pattycake , also known as Patty Cake ( September 3 , 1972 – March 31 , 2013 ) was a female western lowland gorilla ( Gorilla gorilla gorilla ) born to Lulu and Kongo at the Central Park Zoo in New York City . She was the first baby gorilla successfully born in captivity in New York . Months after her much publicized birth , Pattycake 's arm was broken when it got stuck in her cage as her mother grabbed her away from her father . The incident was sensationally anthropomorphized in the media as a domestic dispute between Lulu and Kongo , but in reality experts thought it was a simple accident .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Pattycake , also known as Patty Cake ( September 3 , 1972 – March 31 , 2013 ) was a female western lowland gorilla ( Gorilla gorilla gorilla ) born to Lulu and Kongo at the Central Park Zoo in New York City . She was the first baby gorilla successfully born in captivity in New York . Months after her much publicized birth , Pattycake 's arm was broken when it got stuck in her cage as her mother grabbed her away from her father . The incident was sensationally anthropomorphized in the media as a domestic dispute between Lulu and Kongo , but in reality experts thought it was a simple accident . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Pattycake , also known as Patty Cake ( September 3 , 1972 – March 31 , 2013 ) was<|end-PI|> <|retrieval|> <|start-KFI|>a female western lowland gorilla ( Gorilla gorilla gorilla ) born to Lulu and Kongo at the Central Park Zoo in New York City<|end-KFI|>. <|start-PI|>She was the first baby gorilla successfully<|end-PI|> <|retrieval|> <|start-KFI|>born in captivity in New York<|end-KFI|>. <|start-PI|>Months after her much publicized birth , Pattycake 's arm was broken when<|end-PI|> <|retrieval|> <|start-KFI|>it got stuck in her cage as her mother grabbed her away from her father<|end-KFI|>. <|start-PI|>The incident was sensationally anthropomorphized in the media as<|end-PI|> <|retrieval|> <|start-KFI|>a domestic dispute between Lulu and Kongo , but in reality experts thought it was a simple accident<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces Pattycake and sets up a query about her identity, while the KFI provides specific details about her species and birth. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "Detailed information about Pattycake, including her species, parents, and place of birth, as well as general facts about western lowland gorillas in captivity.",
"Key Factual Information": "a female western lowland gorilla ( Gorilla gorilla gorilla ) born to Lulu and Kongo at the Central Park Zoo in New York City",
"Prefix reformulated": "What type of animal was Pattycake?",
"Prefix-intent": "Pattycake , also known as Patty Cake ( September 3 , 1972 – March 31 , 2013 ) was",
"Verification": "The KFI directly provides information about Pattycake's species and birth details, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about Pattycake's significance, and the KFI provides the specific historical context. The <|retrieval|> token is inserted to indicate where additional information about gorilla births in captivity might be needed.",
"Ideal Knowledge": "Information about the history of gorilla births in captivity, specifically in New York, highlighting Pattycake's status as the first successful birth.",
"Key Factual Information": "born in captivity in New York",
"Prefix reformulated": "What was Pattycake's significance in terms of gorilla births?",
"Prefix-intent": "She was the first baby gorilla successfully",
"Verification": "The KFI directly provides information about Pattycake's historical significance, as prompted by the PI."
},
{
"Explanation": "The PI introduces an incident involving Pattycake's injury, and the KFI provides the specific details of how it occurred. The <|retrieval|> token is inserted to indicate where additional information about the incident might be needed.",
"Ideal Knowledge": "Detailed account of the incident involving Pattycake's arm injury, including information about typical gorilla behavior in captivity and potential risks in zoo environments.",
"Key Factual Information": "it got stuck in her cage as her mother grabbed her away from her father",
"Prefix reformulated": "How did Pattycake break her arm?",
"Prefix-intent": "Months after her much publicized birth , Pattycake 's arm was broken when",
"Verification": "The KFI directly describes the circumstances of Pattycake's arm injury, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the media's portrayal of the incident, and the KFI provides both the sensationalized version and the expert opinion. The <|retrieval|> token is inserted to indicate where additional information about media portrayal versus expert opinion might be needed.",
"Ideal Knowledge": "Information about media sensationalism in reporting animal behavior, contrasted with expert opinions on gorilla behavior in captivity and the nature of such incidents.",
"Key Factual Information": "a domestic dispute between Lulu and Kongo , but in reality experts thought it was a simple accident",
"Prefix reformulated": "How did the media portray the incident of Pattycake's arm injury?",
"Prefix-intent": "The incident was sensationally anthropomorphized in the media as",
"Verification": "The KFI directly contrasts the media portrayal with expert opinion on the incident, as prompted by the PI."
}
]
} |
Her injury was treated at the Bronx Zoo while a custody dispute between the two zoos broke out in public and elicited a range of opinions from experts who believed Pattycake should be returned to her mother . Intense media coverage and public interest brought Pattycake to the attention of a wide audience , with stories focusing on her recovery , her eventual reunion with her parents , and the conditions of zoo animals in Central Park . An ambitious proposal for renovating the Central Park Zoo arose in the wake of the controversy while the zoo received record attendance rates .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Her injury was treated at the Bronx Zoo while a custody dispute between the two zoos broke out in public and elicited a range of opinions from experts who believed Pattycake should be returned to her mother . Intense media coverage and public interest brought Pattycake to the attention of a wide audience , with stories focusing on her recovery , her eventual reunion with her parents , and the conditions of zoo animals in Central Park . An ambitious proposal for renovating the Central Park Zoo arose in the wake of the controversy while the zoo received record attendance rates . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Her injury was treated at the Bronx Zoo while a custody dispute between the two zoos broke out in public and elicited<|end-PI|> <|retrieval|> <|start-KFI|>a range of opinions from experts who believed Pattycake should be returned to her mother<|end-KFI|>. <|start-PI|>Intense media coverage and public interest brought Pattycake to the attention of a wide audience , with stories focusing on<|end-PI|> <|retrieval|> <|start-KFI|>her recovery , her eventual reunion with her parents , and the conditions of zoo animals in Central Park<|end-KFI|>. <|start-PI|>An ambitious proposal for renovating the Central Park Zoo arose in the wake of the controversy while the zoo received<|end-PI|> <|retrieval|> <|start-KFI|>record attendance rates<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of the custody dispute and prompts for its consequences, while the KFI provides specific information about the expert opinions. The <|retrieval|> token is inserted to indicate where additional information about the dispute's impact might be needed.",
"Ideal Knowledge": "Details about the custody dispute over Pattycake, including expert opinions on whether the animal should be returned to its mother, and the reasons behind these opinions.",
"Key Factual Information": "a range of opinions from experts who believed Pattycake should be returned to her mother",
"Prefix reformulated": "What did the custody dispute over Pattycake elicit?",
"Prefix-intent": "Her injury was treated at the Bronx Zoo while a custody dispute between the two zoos broke out in public and elicited",
"Verification": "The KFI directly provides information about the opinions elicited by the custody dispute, as prompted by the PI."
},
{
"Explanation": "The PI introduces the media coverage and sets up a query about its focus, while the KFI provides specific topics that were covered. The <|retrieval|> token is inserted to indicate where additional information about these media stories might be needed.",
"Ideal Knowledge": "Detailed information about Pattycake's recovery process, the reunion with her parents, and an overview of the conditions for zoo animals in Central Park at that time.",
"Key Factual Information": "her recovery , her eventual reunion with her parents , and the conditions of zoo animals in Central Park",
"Prefix reformulated": "What did the media stories about Pattycake focus on?",
"Prefix-intent": "Intense media coverage and public interest brought Pattycake to the attention of a wide audience , with stories focusing on",
"Verification": "The KFI directly lists the topics that the media stories focused on, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of the renovation proposal and prompts for additional consequences, while the KFI provides specific information about the zoo's attendance. The <|retrieval|> token is inserted to indicate where additional information about the zoo's performance might be needed.",
"Ideal Knowledge": "Statistical data on Central Park Zoo's attendance rates before and after the Pattycake controversy, showing the record-breaking numbers and their correlation with the incident.",
"Key Factual Information": "record attendance rates",
"Prefix reformulated": "What did the Central Park Zoo receive after the Pattycake controversy?",
"Prefix-intent": "An ambitious proposal for renovating the Central Park Zoo arose in the wake of the controversy while the zoo received",
"Verification": "The KFI directly provides information about the zoo's attendance rates following the controversy, as prompted by the PI."
}
]
} |
Pattycake was the " child star " of New York City in the early 1970s , and her fame was compared to Shirley Temple . At a time when New York City was facing many problems , she distracted the public from their growing anxieties and became a welcome relief for New Yorkers and their children who loved to visit her . After spending the first decade of her life at the Central Park Zoo , Pattycake moved permanently to the Bronx Zoo in 1982 . She was the mother of ten baby gorillas , including twins born in 1995 . Pattycake spent her later life as an independent but caring troop matriarch in the Bronx Zoo 's Congo Gorilla Forest exhibit . After suffering from arthritis and cardiac problems for some time , Pattycake succumbed to heart disease in 2013 .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Pattycake was the \" child star \" of New York City in the early 1970s , and her fame was compared to Shirley Temple . At a time when New York City was facing many problems , she distracted the public from their growing anxieties and became a welcome relief for New Yorkers and their children who loved to visit her . After spending the first decade of her life at the Central Park Zoo , Pattycake moved permanently to the Bronx Zoo in 1982 . She was the mother of ten baby gorillas , including twins born in 1995 . Pattycake spent her later life as an independent but caring troop matriarch in the Bronx Zoo 's Congo Gorilla Forest exhibit . After suffering from arthritis and cardiac problems for some time , Pattycake succumbed to heart disease in 2013 . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Pattycake was the \" child star \" of New York City in the early 1970s , and her fame was compared to<|end-PI|> <|retrieval|> <|start-KFI|>Shirley Temple<|end-KFI|> . At a time when New York City was facing many problems , she distracted the public from their growing anxieties and became a welcome relief for New Yorkers and their children who loved to visit her . <|start-PI|>After spending the first decade of her life at the Central Park Zoo , Pattycake moved permanently to<|end-PI|> <|retrieval|> <|start-KFI|>the Bronx Zoo in 1982<|end-KFI|> . <|start-PI|>She was the mother of ten baby gorillas , including<|end-PI|> <|retrieval|> <|start-KFI|>twins born in 1995<|end-KFI|> . Pattycake spent her later life as an independent but caring troop matriarch in the Bronx Zoo 's Congo Gorilla Forest exhibit . <|start-PI|>After suffering from arthritis and cardiac problems for some time , Pattycake succumbed to<|end-PI|> <|retrieval|> <|start-KFI|>heart disease in 2013<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces Pattycake as a child star and sets up a comparison, while the KFI provides the specific person she was compared to. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the comparison.",
"Ideal Knowledge": "Information about famous child stars in the 20th century, particularly Shirley Temple, and their impact on popular culture.",
"Key Factual Information": "Shirley Temple",
"Prefix reformulated": "Who was Pattycake's fame compared to in the early 1970s?",
"Prefix-intent": "Pattycake was the \" child star \" of New York City in the early 1970s , and her fame was compared to",
"Verification": "The KFI directly provides the comparison requested in the PI."
},
{
"Explanation": "The PI sets up information about Pattycake's early life and move, while the KFI provides the specific destination and year. The <|retrieval|> token is inserted to indicate where additional information about zoo transfers might be needed.",
"Ideal Knowledge": "Historical information about the transfer of animals between zoos in New York City, specifically focusing on gorilla relocations in the early 1980s.",
"Key Factual Information": "the Bronx Zoo in 1982",
"Prefix reformulated": "Where did Pattycake move to after spending her first decade at the Central Park Zoo?",
"Prefix-intent": "After spending the first decade of her life at the Central Park Zoo , Pattycake moved permanently to",
"Verification": "The KFI directly provides the location and date of Pattycake's move, as prompted by the PI."
},
{
"Explanation": "The PI introduces Pattycake's role as a mother, and the KFI provides specific information about a notable birth. The <|retrieval|> token is inserted to indicate where additional information about gorilla births might be needed.",
"Ideal Knowledge": "Detailed records of gorilla births in captivity, particularly focusing on twin births and their rarity among gorillas.",
"Key Factual Information": "twins born in 1995",
"Prefix reformulated": "What special birth did Pattycake have among her ten baby gorillas?",
"Prefix-intent": "She was the mother of ten baby gorillas , including",
"Verification": "The KFI directly provides information about a special birth among Pattycake's offspring, as prompted by the PI."
},
{
"Explanation": "The PI sets up information about Pattycake's health issues, and the KFI provides the specific cause of death and year. The <|retrieval|> token is inserted to indicate where additional information about gorilla health and longevity might be needed.",
"Ideal Knowledge": "Medical information about common health issues in aging gorillas, particularly focusing on heart disease and its prevalence.",
"Key Factual Information": "heart disease in 2013",
"Prefix reformulated": "What did Pattycake ultimately die from after suffering from arthritis and cardiac problems?",
"Prefix-intent": "After suffering from arthritis and cardiac problems for some time , Pattycake succumbed to",
"Verification": "The KFI directly provides the cause and year of Pattycake's death, as prompted by the PI."
}
]
} |
" Pattycake " was born on September 3 , 1972 , to western lowland gorilla parents Lulu and Kongo at the Central Park Zoo . Lulu and Kongo first arrived at the zoo on May 11 , 1966 . At the time , it was thought that Lulu wasn 't mature enough to conceive , so when Lulu gave birth to Pattycake at the age of 8 , it came as a surprise . Up until the time Lulu gave birth , it had been very difficult to get captive gorilla mothers to raise their young in zoos because the babies would either be rejected by the mother or they would be taken away by the handlers . Pattycake was the first gorilla successfully born in captivity in New York .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n \" Pattycake \" was born on September 3 , 1972 , to western lowland gorilla parents Lulu and Kongo at the Central Park Zoo . Lulu and Kongo first arrived at the zoo on May 11 , 1966 . At the time , it was thought that Lulu wasn 't mature enough to conceive , so when Lulu gave birth to Pattycake at the age of 8 , it came as a surprise . Up until the time Lulu gave birth , it had been very difficult to get captive gorilla mothers to raise their young in zoos because the babies would either be rejected by the mother or they would be taken away by the handlers . Pattycake was the first gorilla successfully born in captivity in New York . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>\"Pattycake\" was born on<|end-PI|> <|retrieval|> <|start-KFI|>September 3, 1972<|end-KFI|>, to western lowland gorilla parents Lulu and Kongo at the Central Park Zoo. <|start-PI|>Lulu and Kongo first arrived at the zoo on<|end-PI|> <|retrieval|> <|start-KFI|>May 11, 1966<|end-KFI|>. At the time, it was thought that Lulu wasn't mature enough to conceive, so when Lulu gave birth to Pattycake at the age of 8, it came as a surprise. Up until the time Lulu gave birth, it had been very difficult to get captive gorilla mothers to raise their young in zoos because the babies would either be rejected by the mother or they would be taken away by the handlers. <|start-PI|>Pattycake was the first gorilla<|end-PI|> <|retrieval|> <|start-KFI|>successfully born in captivity in New York<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces Pattycake and sets up a query about her birth date, while the KFI provides the specific date. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "A detailed record of gorilla births in captivity, specifically mentioning Pattycake's birth date and its significance in the history of captive gorilla breeding.",
"Key Factual Information": "September 3, 1972",
"Prefix reformulated": "When was the gorilla named \"Pattycake\" born?",
"Prefix-intent": "\"Pattycake\" was born on",
"Verification": "The KFI directly provides the birth date requested in the PI."
},
{
"Explanation": "The PI sets up a query about when Lulu and Kongo arrived at the zoo, and the KFI provides the specific date. The <|retrieval|> token is inserted to indicate where additional information about their arrival might be needed.",
"Ideal Knowledge": "A chronological record of gorilla acquisitions by the Central Park Zoo, highlighting the arrival of Lulu and Kongo and their significance to the zoo's gorilla breeding program.",
"Key Factual Information": "May 11, 1966",
"Prefix reformulated": "When did the gorillas Lulu and Kongo arrive at the Central Park Zoo?",
"Prefix-intent": "Lulu and Kongo first arrived at the zoo on",
"Verification": "The KFI directly provides the arrival date of Lulu and Kongo at the zoo, as prompted by the PI."
},
{
"Explanation": "The PI introduces a query about Pattycake's significance, and the KFI provides the specific achievement. The <|retrieval|> token is inserted to indicate where additional information about gorilla breeding in captivity might be needed.",
"Ideal Knowledge": "A comprehensive history of gorilla breeding attempts in New York zoos, highlighting the challenges faced and the significance of Pattycake's successful birth as a milestone in captive gorilla breeding programs.",
"Key Factual Information": "successfully born in captivity in New York",
"Prefix reformulated": "What was significant about Pattycake's birth in relation to gorillas in New York?",
"Prefix-intent": "Pattycake was the first gorilla",
"Verification": "The KFI directly states the significance of Pattycake's birth, as prompted by the PI."
}
]
} |
At the time of her birth , it was unknown if she was a male or female , as it was considered too dangerous to approach her and her parents . Her handlers assumed she was a male and originally named her " Sonny Jim " . When it was established that Pattycake was a female , a contest was held by the New York Daily News to find her a name . " Patty Cake " , the winning entry , was submitted by New York fireman John O 'Connor , who named the gorilla after his wife and a proposed daughter . " It just so happens that we have three boys and I told my wife that if we ever have a girl , we should name her Patty , which happens to be her name . I thought we 'd let the baby gorilla use the name in the meantime , " O 'Connor told reporters .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n At the time of her birth , it was unknown if she was a male or female , as it was considered too dangerous to approach her and her parents . Her handlers assumed she was a male and originally named her \" Sonny Jim \" . When it was established that Pattycake was a female , a contest was held by the New York Daily News to find her a name . \" Patty Cake \" , the winning entry , was submitted by New York fireman John O 'Connor , who named the gorilla after his wife and a proposed daughter . \" It just so happens that we have three boys and I told my wife that if we ever have a girl , we should name her Patty , which happens to be her name . I thought we 'd let the baby gorilla use the name in the meantime , \" O 'Connor told reporters . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>At the time of her birth, it was unknown if she was<|end-PI|> <|retrieval|> <|start-KFI|>a male or female, as it was considered too dangerous to approach her and her parents<|end-KFI|>. <|start-PI|>Her handlers assumed she was a male and originally named her<|end-PI|> <|retrieval|> <|start-KFI|>\"Sonny Jim\"<|end-KFI|>. <|start-PI|>When it was established that Pattycake was a female, a contest was held by<|end-PI|> <|retrieval|> <|start-KFI|>the New York Daily News to find her a name<|end-KFI|>. <|start-PI|>\"Patty Cake\", the winning entry, was submitted by<|end-PI|> <|retrieval|> <|start-KFI|>New York fireman John O'Connor, who named the gorilla after his wife and a proposed daughter<|end-KFI|>. \"It just so happens that we have three boys and I told my wife that if we ever have a girl, we should name her Patty, which happens to be her name. I thought we'd let the baby gorilla use the name in the meantime,\" O'Connor told reporters.",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of Pattycake's birth, while the KFI provides the specific information about the uncertainty of her sex. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "Information about the challenges of determining the sex of newborn gorillas in captivity, including safety concerns for both the baby and its parents.",
"Key Factual Information": "a male or female, as it was considered too dangerous to approach her and her parents",
"Prefix reformulated": "At the time of Pattycake's birth, was it known if she was",
"Prefix-intent": "At the time of her birth, it was unknown if she was",
"Verification": "The KFI directly provides the information about the uncertainty of Pattycake's sex at birth, as prompted by the PI."
},
{
"Explanation": "The PI introduces the handlers' assumption and sets up a query about the original name, while the KFI provides the specific name. The <|retrieval|> token is inserted to indicate where additional information about naming practices might be needed.",
"Ideal Knowledge": "A list of common names given to male gorillas in captivity, including the reasoning behind choosing such names.",
"Key Factual Information": "\"Sonny Jim\"",
"Prefix reformulated": "What did Pattycake's handlers originally name her, assuming she was male?",
"Prefix-intent": "Her handlers assumed she was a male and originally named her",
"Verification": "The KFI directly provides the original name given to Pattycake, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of Pattycake being identified as female and a contest being held, while the KFI specifies who organized the contest. The <|retrieval|> token is inserted to indicate where additional information about the contest might be needed.",
"Ideal Knowledge": "Details about the New York Daily News' involvement in naming contests for zoo animals, including previous examples and the public engagement these contests generate.",
"Key Factual Information": "the New York Daily News to find her a name",
"Prefix reformulated": "Who held a contest to name Pattycake when it was established she was female?",
"Prefix-intent": "When it was established that Pattycake was a female, a contest was held by",
"Verification": "The KFI directly provides information about who held the naming contest, as prompted by the PI."
},
{
"Explanation": "The PI introduces the winning name and sets up a query about its origin, while the KFI provides specific information about the person who submitted it and their motivation. The <|retrieval|> token is inserted to indicate where additional information about John O'Connor might be needed.",
"Ideal Knowledge": "Biographical information about John O'Connor, including his career as a New York fireman and his family background, particularly relating to his wife's name and his desire for a daughter.",
"Key Factual Information": "New York fireman John O'Connor, who named the gorilla after his wife and a proposed daughter",
"Prefix reformulated": "Who submitted the winning entry \"Patty Cake\" in the naming contest?",
"Prefix-intent": "\"Patty Cake\", the winning entry, was submitted by",
"Verification": "The KFI directly provides information about who submitted the winning name and their reasoning, as prompted by the PI."
}
]
} |
Pattycake 's birth caught the attention of the city and brought crowds of thousands of New Yorkers to the Central Park Zoo . Reporter N. R. Kleinfield called her a child star whose " furry face served as a bit of a respite at a time when the city found itself grappling with high crime rates and an intensifying financial crisis . " Six months after Pattycake was born , the director of the zoo estimated that based on the crowds , she might draw an additional 500 @,@ 000 visitors by the time of her first birthday .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Pattycake 's birth caught the attention of the city and brought crowds of thousands of New Yorkers to the Central Park Zoo . Reporter N. R. Kleinfield called her a child star whose \" furry face served as a bit of a respite at a time when the city found itself grappling with high crime rates and an intensifying financial crisis . \" Six months after Pattycake was born , the director of the zoo estimated that based on the crowds , she might draw an additional 500 @,@ 000 visitors by the time of her first birthday . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Pattycake's birth caught the attention of the city and brought<|end-PI|> <|retrieval|> <|start-KFI|>crowds of thousands of New Yorkers to the Central Park Zoo<|end-KFI|>. <|start-PI|>Reporter N. R. Kleinfield called her a child star whose<|end-PI|> <|retrieval|> <|start-KFI|>\"furry face served as a bit of a respite at a time when the city found itself grappling with high crime rates and an intensifying financial crisis.\"<|end-KFI|> <|start-PI|>Six months after Pattycake was born, the director of the zoo estimated that based on the crowds, she might draw<|end-PI|> <|retrieval|> <|start-KFI|>an additional 500,000 visitors by the time of her first birthday<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces Pattycake's birth and its impact, while the KFI specifies the scale of public interest. The <|retrieval|> token is inserted to indicate where additional information about the event's popularity might be needed.",
"Ideal Knowledge": "Detailed information about Pattycake's birth at the Central Park Zoo, including visitor statistics and the public's reaction to the event.",
"Key Factual Information": "crowds of thousands of New Yorkers to the Central Park Zoo",
"Prefix reformulated": "How many people did Pattycake's birth bring to the Central Park Zoo?",
"Prefix-intent": "Pattycake's birth caught the attention of the city and brought",
"Verification": "The KFI directly provides information about the crowds attracted by Pattycake's birth, as prompted by the PI."
},
{
"Explanation": "The PI introduces the reporter's perspective, while the KFI provides the specific quote describing Pattycake's significance. The <|retrieval|> token is inserted to indicate where additional information about the reporter's views might be needed.",
"Ideal Knowledge": "Quotes and analysis from N. R. Kleinfield's reporting on Pattycake, including context about New York City's social and economic challenges at the time.",
"Key Factual Information": "\"furry face served as a bit of a respite at a time when the city found itself grappling with high crime rates and an intensifying financial crisis.\"",
"Prefix reformulated": "How did Reporter N. R. Kleinfield describe Pattycake's impact?",
"Prefix-intent": "Reporter N. R. Kleinfield called her a child star whose",
"Verification": "The KFI directly provides the reporter's description of Pattycake's impact, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context for the zoo director's estimate, while the KFI provides the specific projected number. The <|retrieval|> token is inserted to indicate where additional information about visitor statistics might be needed.",
"Ideal Knowledge": "Statistical data and projections from the Central Park Zoo regarding visitor numbers before and after Pattycake's birth, including the director's estimates and methodology.",
"Key Factual Information": "an additional 500,000 visitors by the time of her first birthday",
"Prefix reformulated": "How many additional visitors did the zoo director estimate Pattycake would attract by her first birthday?",
"Prefix-intent": "Six months after Pattycake was born, the director of the zoo estimated that based on the crowds, she might draw",
"Verification": "The KFI directly provides the estimated number of additional visitors, as prompted by the PI."
}
]
} |
At the age of five months , Pattycake was sharing the Lion House with Lulu and Kongo at the Central Park Zoo . On March 20 , 1973 , an accident occurred that was attributed more to clumsiness than to parenting skills . Although no staff were on hand to see it directly , according to visitors , while little Pattycake was reaching out to her father through the bars of her cage , her right arm became stuck somehow when her mother Lulu pulled her away , breaking her right arm . Zoo handlers had to subdue Lulu with a tranquilizer dart to remove Pattycake and treat her injury . While experts considered the incident an accident , the media sensationalized the event . Zookeeper Veronica Nelson , who worked with Patty Cake , recalled that
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n At the age of five months , Pattycake was sharing the Lion House with Lulu and Kongo at the Central Park Zoo . On March 20 , 1973 , an accident occurred that was attributed more to clumsiness than to parenting skills . Although no staff were on hand to see it directly , according to visitors , while little Pattycake was reaching out to her father through the bars of her cage , her right arm became stuck somehow when her mother Lulu pulled her away , breaking her right arm . Zoo handlers had to subdue Lulu with a tranquilizer dart to remove Pattycake and treat her injury . While experts considered the incident an accident , the media sensationalized the event . Zookeeper Veronica Nelson , who worked with Patty Cake , recalled that \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "At the age of five months, <|start-PI|>Pattycake was sharing the Lion House with<|end-PI|> <|retrieval|> <|start-KFI|>Lulu and Kongo at the Central Park Zoo<|end-KFI|>. <|start-PI|>On March 20, 1973, an accident occurred that was attributed more to<|end-PI|> <|retrieval|> <|start-KFI|>clumsiness than to parenting skills<|end-KFI|>. Although no staff were on hand to see it directly, <|start-PI|>according to visitors, while little Pattycake was reaching out to her father through the bars of her cage, her right arm became stuck somehow when<|end-PI|> <|retrieval|> <|start-KFI|>her mother Lulu pulled her away, breaking her right arm<|end-KFI|>. <|start-PI|>Zoo handlers had to subdue Lulu with<|end-PI|> <|retrieval|> <|start-KFI|>a tranquilizer dart to remove Pattycake and treat her injury<|end-KFI|>. While experts considered the incident an accident, the media sensationalized the event. Zookeeper Veronica Nelson, who worked with Patty Cake, recalled that",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context about Pattycake's living situation, and the KFI provides the specific details about her companions and location. The <|retrieval|> token is inserted to indicate where additional information about the zoo inhabitants might be needed.",
"Ideal Knowledge": "Information about the inhabitants of the Lion House at the Central Park Zoo, specifically mentioning Pattycake, Lulu, and Kongo, and their relationships.",
"Key Factual Information": "Lulu and Kongo at the Central Park Zoo",
"Prefix reformulated": "Who was Pattycake sharing the Lion House with at the age of five months?",
"Prefix-intent": "At the age of five months, Pattycake was sharing the Lion House with",
"Verification": "The KFI directly provides the information about who Pattycake was sharing the Lion House with, as prompted by the PI."
},
{
"Explanation": "The PI introduces the accident and sets up a query about its cause, while the KFI provides the specific attribution. The <|retrieval|> token is inserted to indicate where additional information about the accident might be needed.",
"Ideal Knowledge": "Detailed account of the accident involving Pattycake on March 20, 1973, including expert opinions on the cause and nature of the incident.",
"Key Factual Information": "clumsiness than to parenting skills",
"Prefix reformulated": "What was the accident on March 20, 1973 attributed to?",
"Prefix-intent": "On March 20, 1973, an accident occurred that was attributed more to",
"Verification": "The KFI directly provides the attribution of the accident, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of the accident, and the KFI provides the specific details about how Pattycake's arm was broken. The <|retrieval|> token is inserted to indicate where additional information about the accident might be needed.",
"Ideal Knowledge": "Detailed eyewitness accounts of the accident involving Pattycake, including the sequence of events leading to her injury and the roles of her parents in the incident.",
"Key Factual Information": "her mother Lulu pulled her away, breaking her right arm",
"Prefix reformulated": "What happened to Pattycake's right arm when she was reaching out to her father?",
"Prefix-intent": "According to visitors, while little Pattycake was reaching out to her father through the bars of her cage, her right arm became stuck somehow when",
"Verification": "The KFI directly describes what happened to Pattycake's arm, as prompted by the PI."
},
{
"Explanation": "The PI introduces the zoo handlers' action, and the KFI provides the specific method used and its purpose. The <|retrieval|> token is inserted to indicate where additional information about zoo emergency procedures might be needed.",
"Ideal Knowledge": "Information on zoo protocols for handling emergencies involving animals, specifically the use of tranquilizer darts and procedures for treating injured animals.",
"Key Factual Information": "a tranquilizer dart to remove Pattycake and treat her injury",
"Prefix reformulated": "How did zoo handlers subdue Lulu?",
"Prefix-intent": "Zoo handlers had to subdue Lulu with",
"Verification": "The KFI directly provides information on how Lulu was subdued and why, as prompted by the PI."
}
]
} |
The news media would have liked to have it a dramatic bloody mess — a struggle between mom and pop for the custody of the child . It was nothing like that . It was a simple accident . Kongo was in one part of the double cage , Lulu in another , and between them was a partition of narrow bars . Lulu had Patty Cake in her arms and when Patty reached in between the narrow bars to touch her father , Lulu suddenly pulled her away . But Patty 's arm got caught in the narrow bars and broke . It was a freak accident . No one ever realized that those bars were narrow enough to catch that tiny arm .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n The news media would have liked to have it a dramatic bloody mess — a struggle between mom and pop for the custody of the child . It was nothing like that . It was a simple accident . Kongo was in one part of the double cage , Lulu in another , and between them was a partition of narrow bars . Lulu had Patty Cake in her arms and when Patty reached in between the narrow bars to touch her father , Lulu suddenly pulled her away . But Patty 's arm got caught in the narrow bars and broke . It was a freak accident . No one ever realized that those bars were narrow enough to catch that tiny arm . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
Pattycake was brought to New York Medical College for surgery and she was given a cast for her arm . Due to concerns that Lulu would try to remove Pattycake 's cast , she was separated from her mother and moved to the Bronx Zoo for convalescence . Pattycake was treated by veterinarian Emil Dolensek who later replaced her cast with a sling . After an examination , the staff discovered that Pattycake had intestinal parasites and determined she was underfed . They also believed that as a result of the incident , Lulu wasn 't capable as a mother .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Pattycake was brought to New York Medical College for surgery and she was given a cast for her arm . Due to concerns that Lulu would try to remove Pattycake 's cast , she was separated from her mother and moved to the Bronx Zoo for convalescence . Pattycake was treated by veterinarian Emil Dolensek who later replaced her cast with a sling . After an examination , the staff discovered that Pattycake had intestinal parasites and determined she was underfed . They also believed that as a result of the incident , Lulu wasn 't capable as a mother . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
A custody dispute began between the two zoos , with the Bronx Zoo arguing that she would be better cared for in their facilities . Time magazine noted that it was the " custody battle of the decade " in the " primate world " , comparing Patty Cake 's popularity and fame to that of child star Shirley Temple . Developmental biologist Ronald Nadler of the Yerkes National Primate Research Center was brought in to arbitrate the dispute and published a report that favored returning Pattycake to her mother and the Central Park Zoo . In his report Nadler noted that " the recommendation is based on the judgment that an infant gorilla is more likely to develop into a socially competent and reproductively adequate animal if it is raised in the company of its parents as opposed to being raised with a group of peers . "
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n A custody dispute began between the two zoos , with the Bronx Zoo arguing that she would be better cared for in their facilities . Time magazine noted that it was the \" custody battle of the decade \" in the \" primate world \" , comparing Patty Cake 's popularity and fame to that of child star Shirley Temple . Developmental biologist Ronald Nadler of the Yerkes National Primate Research Center was brought in to arbitrate the dispute and published a report that favored returning Pattycake to her mother and the Central Park Zoo . In his report Nadler noted that \" the recommendation is based on the judgment that an infant gorilla is more likely to develop into a socially competent and reproductively adequate animal if it is raised in the company of its parents as opposed to being raised with a group of peers . \" \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
After three months recuperating from her injuries , Pattycake was returned to her mother on June 15 , 1973 . The entire incident was documented by artist Susan Green in her book Gentle Gorilla : The Story of Patty Cake ( 1978 ) .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n After three months recuperating from her injuries , Pattycake was returned to her mother on June 15 , 1973 . The entire incident was documented by artist Susan Green in her book Gentle Gorilla : The Story of Patty Cake ( 1978 ) . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
By the 1970s , animal welfare organizations began to voice their increasing concern with the treatment of zoo animals and the conditions of their enclosures at the Central Park Zoo , the oldest zoo in the U.S. The Humane Society of the United States , the Friends of the Zoo , and the Society for Animal Rights decried the prison @-@ like conditions of the cages and called for changes . The New York Zoological Society , which was responsible for creating realistic habitat enclosures at the Bronx Zoo , also began calling for changes and for Pattycake to be moved to another zoo . A renovation plan for the Central Park Zoo was approved in 1981 , with plans made to move the gorillas to larger spaces in other zoos . The New York Times reported that " the caging of these animals in inadequate spaces has long enraged animal lovers . " When the zoo was finally closed for renovations in 1982 , Pattycake was moved to the Bronx Zoo .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n By the 1970s , animal welfare organizations began to voice their increasing concern with the treatment of zoo animals and the conditions of their enclosures at the Central Park Zoo , the oldest zoo in the U.S. The Humane Society of the United States , the Friends of the Zoo , and the Society for Animal Rights decried the prison @-@ like conditions of the cages and called for changes . The New York Zoological Society , which was responsible for creating realistic habitat enclosures at the Bronx Zoo , also began calling for changes and for Pattycake to be moved to another zoo . A renovation plan for the Central Park Zoo was approved in 1981 , with plans made to move the gorillas to larger spaces in other zoos . The New York Times reported that \" the caging of these animals in inadequate spaces has long enraged animal lovers . \" When the zoo was finally closed for renovations in 1982 , Pattycake was moved to the Bronx Zoo . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "By the 1970s, <|start-PI|>animal welfare organizations began to voice their increasing concern with the treatment of zoo animals and the conditions of their enclosures at<|end-PI|> <|retrieval|> <|start-KFI|>the Central Park Zoo, the oldest zoo in the U.S.<|end-KFI|> <|start-PI|>The Humane Society of the United States, the Friends of the Zoo, and the Society for Animal Rights decried<|end-PI|> <|retrieval|> <|start-KFI|>the prison-like conditions of the cages and called for changes<|end-KFI|>. The New York Zoological Society, which was responsible for creating realistic habitat enclosures at the Bronx Zoo, also began calling for changes and for Pattycake to be moved to another zoo. <|start-PI|>A renovation plan for the Central Park Zoo was approved in<|end-PI|> <|retrieval|> <|start-KFI|>1981, with plans made to move the gorillas to larger spaces in other zoos<|end-KFI|>. <|start-PI|>The New York Times reported that<|end-PI|> <|retrieval|> <|start-KFI|>\"the caging of these animals in inadequate spaces has long enraged animal lovers.\"<|end-KFI|> <|start-PI|>When the zoo was finally closed for renovations in 1982,<|end-PI|> <|retrieval|> <|start-KFI|>Pattycake was moved to the Bronx Zoo<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of animal welfare concerns in the 1970s, while the KFI specifies the zoo in question. The <|retrieval|> token is inserted to indicate where additional information about the Central Park Zoo might be needed.",
"Ideal Knowledge": "Information about the Central Park Zoo's history, its status as the oldest zoo in the U.S., and details about animal welfare concerns raised in the 1970s.",
"Key Factual Information": "the Central Park Zoo, the oldest zoo in the U.S.",
"Prefix reformulated": "Which zoo was criticized for its animal treatment and enclosure conditions in the 1970s?",
"Prefix-intent": "By the 1970s, animal welfare organizations began to voice their increasing concern with the treatment of zoo animals and the conditions of their enclosures at",
"Verification": "The KFI directly identifies the zoo mentioned in the PI and provides additional context."
},
{
"Explanation": "The PI lists the organizations involved, and the KFI provides the specific nature of their criticism. The <|retrieval|> token is inserted to indicate where additional information about the organizations' concerns might be needed.",
"Ideal Knowledge": "Detailed accounts of the specific criticisms made by these animal rights organizations regarding the Central Park Zoo's enclosures, including comparisons to prison conditions.",
"Key Factual Information": "the prison-like conditions of the cages and called for changes",
"Prefix reformulated": "What did animal rights organizations criticize about the Central Park Zoo?",
"Prefix-intent": "The Humane Society of the United States, the Friends of the Zoo, and the Society for Animal Rights decried",
"Verification": "The KFI directly states the criticism and demands made by the organizations mentioned in the PI."
},
{
"Explanation": "The PI introduces the renovation plan, and the KFI provides the specific year and details. The <|retrieval|> token is inserted to indicate where additional information about the renovation plan might be needed.",
"Ideal Knowledge": "Detailed information about the 1981 Central Park Zoo renovation plan, including specific provisions for relocating gorillas and improving animal habitats.",
"Key Factual Information": "1981, with plans made to move the gorillas to larger spaces in other zoos",
"Prefix reformulated": "When was the renovation plan for the Central Park Zoo approved?",
"Prefix-intent": "A renovation plan for the Central Park Zoo was approved in",
"Verification": "The KFI directly provides the year of approval and additional details about the renovation plans, as prompted by the PI."
},
{
"Explanation": "The PI introduces the New York Times report, and the KFI provides the direct quote. The <|retrieval|> token is inserted to indicate where the full report or additional context might be needed.",
"Ideal Knowledge": "The full New York Times article or report discussing the Central Park Zoo's animal enclosures, including context and additional quotes from animal welfare advocates.",
"Key Factual Information": "\"the caging of these animals in inadequate spaces has long enraged animal lovers.\"",
"Prefix reformulated": "What did The New York Times report about the Central Park Zoo's animal enclosures?",
"Prefix-intent": "The New York Times reported that",
"Verification": "The KFI directly quotes the New York Times report, providing the exact statement as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of the zoo's closure, and the KFI provides information about what happened to Pattycake. The <|retrieval|> token is inserted to indicate where additional information about the zoo's closure and animal relocation might be needed.",
"Ideal Knowledge": "Detailed information about the closure of the Central Park Zoo in 1982, the relocation process for its animals, and specific details about Pattycake's move to the Bronx Zoo.",
"Key Factual Information": "Pattycake was moved to the Bronx Zoo",
"Prefix reformulated": "What happened to Pattycake when the Central Park Zoo closed for renovations in 1982?",
"Prefix-intent": "When the zoo was finally closed for renovations in 1982,",
"Verification": "The KFI directly states what happened to Pattycake when the zoo closed, as prompted by the PI."
}
]
} |
Pattycake moved permanently to the Bronx Zoo on December 20 , 1982 . For a few years , she lived in a cage with Pansy , a chimpanzee . In June 1999 , Pattycake moved into the Wildlife Conservation Society 's $ 43 million Congo Gorilla Forest exhibit . The exhibit includes a Great Gorilla Forest viewing area that separates gorillas and visitors with a glass window . Two troops of gorillas inhabited the 6 @.@ 5 acre exhibit , with a dozen gorillas in Pattycake 's troop alone , including Fubo , Pattycake , Tunko , Triska , Halima , Fran , Layla , Kumi , Suki , Babatunde , Barbara , and M 'domo . The general curator of the Bronx Zoo , James Doherty , described Pattycake as " independent " with " few close friends " in the Congo Gorilla Forest . " It may have something to do with the fact that she didn 't live with her parents that long , and lived with that chimpanzee for a few years , " Doherty said .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Pattycake moved permanently to the Bronx Zoo on December 20 , 1982 . For a few years , she lived in a cage with Pansy , a chimpanzee . In June 1999 , Pattycake moved into the Wildlife Conservation Society 's $ 43 million Congo Gorilla Forest exhibit . The exhibit includes a Great Gorilla Forest viewing area that separates gorillas and visitors with a glass window . Two troops of gorillas inhabited the 6 @.@ 5 acre exhibit , with a dozen gorillas in Pattycake 's troop alone , including Fubo , Pattycake , Tunko , Triska , Halima , Fran , Layla , Kumi , Suki , Babatunde , Barbara , and M 'domo . The general curator of the Bronx Zoo , James Doherty , described Pattycake as \" independent \" with \" few close friends \" in the Congo Gorilla Forest . \" It may have something to do with the fact that she didn 't live with her parents that long , and lived with that chimpanzee for a few years , \" Doherty said . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
Pattycake gave birth to her first baby , Tumai , a male gorilla , on January 20 , 1985 . Tumai was sired by Bendera . Pattycake and Bendera had a second baby on March 23 , 1986 , but it died soon after birth . With Barney , Pattycake gave birth to three babies : a female named Paki , on May 26 , 1989 , followed by Patrick on April 19 , 1990 , and Husani on December 14 , 1991 . Paki gave birth to Pattycake 's only grandchild , Pendeka , in 1998 .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Pattycake gave birth to her first baby , Tumai , a male gorilla , on January 20 , 1985 . Tumai was sired by Bendera . Pattycake and Bendera had a second baby on March 23 , 1986 , but it died soon after birth . With Barney , Pattycake gave birth to three babies : a female named Paki , on May 26 , 1989 , followed by Patrick on April 19 , 1990 , and Husani on December 14 , 1991 . Paki gave birth to Pattycake 's only grandchild , Pendeka , in 1998 . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Pattycake gave birth to her first baby , Tumai , a male gorilla , on<|end-PI|> <|retrieval|> <|start-KFI|>January 20 , 1985<|end-KFI|>. <|start-PI|>Tumai was sired by<|end-PI|> <|retrieval|> <|start-KFI|>Bendera<|end-KFI|>. <|start-PI|>Pattycake and Bendera had a second baby on<|end-PI|> <|retrieval|> <|start-KFI|>March 23 , 1986 , but it died soon after birth<|end-KFI|>. <|start-PI|>With Barney , Pattycake gave birth to three babies :<|end-PI|> <|retrieval|> <|start-KFI|>a female named Paki , on May 26 , 1989 , followed by Patrick on April 19 , 1990 , and Husani on December 14 , 1991<|end-KFI|>. <|start-PI|>Paki gave birth to Pattycake 's only grandchild , Pendeka , in<|end-PI|> <|retrieval|> <|start-KFI|>1998<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces Pattycake's first baby and sets up a query about the birth date, while the KFI provides the specific date. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "A detailed record of gorilla births in captivity, specifically mentioning Pattycake and her offspring, with exact dates and names.",
"Key Factual Information": "January 20 , 1985",
"Prefix reformulated": "When did Pattycake give birth to her first baby, Tumai?",
"Prefix-intent": "Pattycake gave birth to her first baby , Tumai , a male gorilla , on",
"Verification": "The KFI directly provides the date of birth for Tumai, as requested in the PI."
},
{
"Explanation": "The PI sets up a query about Tumai's father, and the KFI provides the specific name. The <|retrieval|> token is inserted to indicate where additional information about gorilla lineage might be needed.",
"Ideal Knowledge": "Genealogical information about captive gorillas, including details about Tumai's parentage and Bendera's role as the father.",
"Key Factual Information": "Bendera",
"Prefix reformulated": "Who was Tumai's father?",
"Prefix-intent": "Tumai was sired by",
"Verification": "The KFI directly provides the name of Tumai's father, as prompted by the PI."
},
{
"Explanation": "The PI introduces the second baby and sets up a query about its birth date, while the KFI provides the specific date and additional information about its fate. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "Detailed records of gorilla births and infant mortality in captivity, specifically mentioning Pattycake and Bendera's second offspring.",
"Key Factual Information": "March 23 , 1986 , but it died soon after birth",
"Prefix reformulated": "When did Pattycake and Bendera have their second baby?",
"Prefix-intent": "Pattycake and Bendera had a second baby on",
"Verification": "The KFI directly provides the date of birth for the second baby and its fate, as prompted by the PI."
},
{
"Explanation": "The PI introduces Pattycake's babies with Barney and sets up a query about their identities, while the KFI provides the specific names and birth dates. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "A comprehensive list of Pattycake's offspring, including names, genders, and exact birth dates, specifically focusing on those fathered by Barney.",
"Key Factual Information": "a female named Paki , on May 26 , 1989 , followed by Patrick on April 19 , 1990 , and Husani on December 14 , 1991",
"Prefix reformulated": "Who were the three babies Pattycake had with Barney?",
"Prefix-intent": "With Barney , Pattycake gave birth to three babies :",
"Verification": "The KFI directly provides the names and birth dates of the three babies, as requested in the PI."
},
{
"Explanation": "The PI introduces Pattycake's grandchild and sets up a query about the birth year, while the KFI provides the specific year. The <|retrieval|> token is inserted to indicate where additional information about gorilla family trees might be needed.",
"Ideal Knowledge": "Genealogical information about Pattycake's extended family, including details about her grandchildren, specifically mentioning Paki's offspring Pendeka and the year of birth.",
"Key Factual Information": "1998",
"Prefix reformulated": "When did Paki give birth to Pattycake's only grandchild, Pendeka?",
"Prefix-intent": "Paki gave birth to Pattycake 's only grandchild , Pendeka , in",
"Verification": "The KFI directly provides the year of birth for Pendeka, as prompted by the PI."
}
]
} |
In October 1991 , a silverback named Timmy ( 1959 – 2011 ) was taken away from Kate , his infertile companion at the Cleveland Metroparks Zoo , in the hopes of breeding lowland gorillas and introducing new genes into the captive gorilla gene pool . This forced separation led to protests from animal rights activists who expressed concerns about the potential consequences of emotional trauma on the two gorillas . The Association of Zoos and Aquariums Species Survival Plan sent Timmy to the Bronx Zoo where he joined Pattycake and other females .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n In October 1991 , a silverback named Timmy ( 1959 – 2011 ) was taken away from Kate , his infertile companion at the Cleveland Metroparks Zoo , in the hopes of breeding lowland gorillas and introducing new genes into the captive gorilla gene pool . This forced separation led to protests from animal rights activists who expressed concerns about the potential consequences of emotional trauma on the two gorillas . The Association of Zoos and Aquariums Species Survival Plan sent Timmy to the Bronx Zoo where he joined Pattycake and other females . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
On July 11 , 1993 , Pattycake and Timmy gave birth to Okpara , a male gorilla . The pair also gave birth to twin males , Ngoma and Tambo , on August 8 , 1994 . It was the sixth time western lowland gorillas had given birth to twins in captivity . The twins were raised in a separate habitat by surrogate mothers . After seven years , Ngoma and Tambo , along with another gorilla named Dan , left for a zoo in Nebraska in February 2001 . On February 4 , 2001 , Pattycake and Zuri gave birth to Dossi , a female . Her last and tenth baby gorilla , a male , was delivered on April 15 , 2002 . Unnamed , it died four days later .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n On July 11 , 1993 , Pattycake and Timmy gave birth to Okpara , a male gorilla . The pair also gave birth to twin males , Ngoma and Tambo , on August 8 , 1994 . It was the sixth time western lowland gorillas had given birth to twins in captivity . The twins were raised in a separate habitat by surrogate mothers . After seven years , Ngoma and Tambo , along with another gorilla named Dan , left for a zoo in Nebraska in February 2001 . On February 4 , 2001 , Pattycake and Zuri gave birth to Dossi , a female . Her last and tenth baby gorilla , a male , was delivered on April 15 , 2002 . Unnamed , it died four days later . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>On July 11 , 1993 , Pattycake and Timmy gave birth to<|end-PI|> <|retrieval|> <|start-KFI|>Okpara , a male gorilla<|end-KFI|> . <|start-PI|>The pair also gave birth to twin males , Ngoma and Tambo , on<|end-PI|> <|retrieval|> <|start-KFI|>August 8 , 1994<|end-KFI|> . <|start-PI|>It was the sixth time western lowland gorillas had given birth to twins in<|end-PI|> <|retrieval|> <|start-KFI|>captivity<|end-KFI|> . The twins were raised in a separate habitat by surrogate mothers . <|start-PI|>After seven years , Ngoma and Tambo , along with another gorilla named Dan , left for a zoo in<|end-PI|> <|retrieval|> <|start-KFI|>Nebraska in February 2001<|end-KFI|> . <|start-PI|>On February 4 , 2001 , Pattycake and Zuri gave birth to<|end-PI|> <|retrieval|> <|start-KFI|>Dossi , a female<|end-KFI|> . <|start-PI|>Her last and tenth baby gorilla , a male , was delivered on<|end-PI|> <|retrieval|> <|start-KFI|>April 15 , 2002<|end-KFI|> . Unnamed , it died four days later .",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of a birth event with specific parents and date, while the KFI provides the name and gender of the offspring. The <|retrieval|> token is inserted to indicate where additional information about the birth might be needed.",
"Ideal Knowledge": "A detailed record of gorilla births in captivity, specifically mentioning Pattycake and Timmy's offspring born on July 11, 1993, including the name and gender of the baby gorilla.",
"Key Factual Information": "Okpara , a male gorilla",
"Prefix reformulated": "Who did Pattycake and Timmy give birth to on July 11, 1993?",
"Prefix-intent": "On July 11 , 1993 , Pattycake and Timmy gave birth to",
"Verification": "The KFI directly provides the name and gender of the offspring born on the specified date, as prompted by the PI."
},
{
"Explanation": "The PI introduces the birth of twin males with their names, and the KFI provides the specific date. The <|retrieval|> token is inserted to indicate where additional information about this birth event might be needed.",
"Ideal Knowledge": "A chronological list of notable gorilla births in captivity, highlighting the birth of twin males Ngoma and Tambo to Pattycake and Timmy on August 8, 1994.",
"Key Factual Information": "August 8 , 1994",
"Prefix reformulated": "When did Pattycake and Timmy give birth to twin males Ngoma and Tambo?",
"Prefix-intent": "The pair also gave birth to twin males , Ngoma and Tambo , on",
"Verification": "The KFI directly provides the date of birth for the twin males, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of the rarity of twin births among western lowland gorillas, and the KFI specifies that this occurred in captivity. The <|retrieval|> token is inserted to indicate where additional information about gorilla twin births might be needed.",
"Ideal Knowledge": "A comprehensive record of twin births among western lowland gorillas, both in the wild and in captivity, highlighting the rarity of such events and noting this as the sixth occurrence in a captive setting.",
"Key Factual Information": "captivity",
"Prefix reformulated": "Where had western lowland gorillas given birth to twins for the sixth time?",
"Prefix-intent": "It was the sixth time western lowland gorillas had given birth to twins in",
"Verification": "The KFI directly answers where the sixth occurrence of twin births for western lowland gorillas took place, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of the gorillas' move after a specific time period, and the KFI provides the destination and date. The <|retrieval|> token is inserted to indicate where additional information about this transfer might be needed.",
"Ideal Knowledge": "Information on gorilla transfers between zoos, specifically detailing the move of Ngoma, Tambo, and Dan to a zoo in Nebraska in February 2001, including the reasons for the transfer and the receiving institution.",
"Key Factual Information": "Nebraska in February 2001",
"Prefix reformulated": "Where did Ngoma, Tambo, and Dan move to after seven years?",
"Prefix-intent": "After seven years , Ngoma and Tambo , along with another gorilla named Dan , left for a zoo in",
"Verification": "The KFI directly provides the location and date of the gorillas' move, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of a birth event with specific parents and date, while the KFI provides the name and gender of the offspring. The <|retrieval|> token is inserted to indicate where additional information about this birth might be needed.",
"Ideal Knowledge": "A detailed record of gorilla births in captivity, specifically mentioning Pattycake and Zuri's offspring born on February 4, 2001, including the name and gender of the baby gorilla.",
"Key Factual Information": "Dossi , a female",
"Prefix reformulated": "Who did Pattycake and Zuri give birth to on February 4, 2001?",
"Prefix-intent": "On February 4 , 2001 , Pattycake and Zuri gave birth to",
"Verification": "The KFI directly provides the name and gender of the offspring born on the specified date, as prompted by the PI."
},
{
"Explanation": "The PI introduces the context of Pattycake's last baby, and the KFI provides the specific date of birth. The <|retrieval|> token is inserted to indicate where additional information about this final birth might be needed.",
"Ideal Knowledge": "A complete birth record for Pattycake, highlighting her tenth and final offspring, born on April 15, 2002, including details about the pregnancy and any complications.",
"Key Factual Information": "April 15 , 2002",
"Prefix reformulated": "When was Pattycake's last and tenth baby gorilla delivered?",
"Prefix-intent": "Her last and tenth baby gorilla , a male , was delivered on",
"Verification": "The KFI directly provides the date of birth for Pattycake's last and tenth baby gorilla, as prompted by the PI."
}
]
} |
Kongo , Pattycake 's father , died in 1998 . In 2002 , the Pattycake Fund was established to raise $ 250 @,@ 000 to stop illegal poaching of African gorillas . The fundraising coincided with Pattycake 's 30th birthday , commemorated with a two @-@ day celebration at the Bronx Zoo . A special cake was made for her from kale , gelatin , yogurt , and berries .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Kongo , Pattycake 's father , died in 1998 . In 2002 , the Pattycake Fund was established to raise $ 250 @,@ 000 to stop illegal poaching of African gorillas . The fundraising coincided with Pattycake 's 30th birthday , commemorated with a two @-@ day celebration at the Bronx Zoo . A special cake was made for her from kale , gelatin , yogurt , and berries . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Kongo, Pattycake's father, died in<|end-PI|> <|retrieval|> <|start-KFI|>1998<|end-KFI|>. <|start-PI|>In 2002, the Pattycake Fund was established to raise<|end-PI|> <|retrieval|> <|start-KFI|>$ 250,000 to stop illegal poaching of African gorillas<|end-KFI|>. <|start-PI|>The fundraising coincided with Pattycake's 30th birthday, commemorated with<|end-PI|> <|retrieval|> <|start-KFI|>a two-day celebration at the Bronx Zoo<|end-KFI|>. <|start-PI|>A special cake was made for her from<|end-PI|> <|retrieval|> <|start-KFI|>kale, gelatin, yogurt, and berries<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces Kongo and sets up a query about his death, while the KFI provides the specific year. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "A timeline of significant events in the life of Pattycake the gorilla, including the death of her father Kongo in 1998.",
"Key Factual Information": "1998",
"Prefix reformulated": "When did Kongo, Pattycake's father, die?",
"Prefix-intent": "Kongo, Pattycake's father, died in",
"Verification": "The KFI directly provides the year of Kongo's death, as prompted by the PI."
},
{
"Explanation": "The PI introduces the Pattycake Fund and sets up a query about its fundraising goal, while the KFI provides the specific amount and purpose. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "Detailed information about the Pattycake Fund, including its establishment date, fundraising goal, and purpose related to gorilla conservation efforts.",
"Key Factual Information": "$ 250,000 to stop illegal poaching of African gorillas",
"Prefix reformulated": "How much money was the Pattycake Fund established to raise in 2002?",
"Prefix-intent": "In 2002, the Pattycake Fund was established to raise",
"Verification": "The KFI directly provides the amount of money and purpose of the fund, as prompted by the PI."
},
{
"Explanation": "The PI mentions Pattycake's birthday and sets up a query about its commemoration, while the KFI provides specific details about the celebration. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.",
"Ideal Knowledge": "Information about Pattycake's 30th birthday celebration, including details about the event's duration and location at the Bronx Zoo.",
"Key Factual Information": "a two-day celebration at the Bronx Zoo",
"Prefix reformulated": "How was Pattycake's 30th birthday commemorated?",
"Prefix-intent": "The fundraising coincided with Pattycake's 30th birthday, commemorated with",
"Verification": "The KFI directly describes how Pattycake's birthday was celebrated, as prompted by the PI."
},
{
"Explanation": "The PI introduces the special cake and sets up a query about its ingredients, while the KFI provides the specific list of ingredients. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "Information about dietary preferences and special treats for gorillas in captivity, specifically mentioning the ingredients used in Pattycake's birthday cake.",
"Key Factual Information": "kale, gelatin, yogurt, and berries",
"Prefix reformulated": "What ingredients were used to make Pattycake's special cake?",
"Prefix-intent": "A special cake was made for her from",
"Verification": "The KFI directly lists the ingredients used in Pattycake's cake, as prompted by the PI."
}
]
} |
Timmy , along with two female gorillas , Tunuka and Paki , left the Bronx Zoo for the Louisville Zoo in May 2004 .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Timmy , along with two female gorillas , Tunuka and Paki , left the Bronx Zoo for the Louisville Zoo in May 2004 . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Timmy , along with two female gorillas , Tunuka and Paki , left the Bronx Zoo for<|end-PI|> <|retrieval|> <|start-KFI|>the Louisville Zoo in May 2004<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of the gorillas leaving the Bronx Zoo, while the KFI provides the specific destination and date. The <|retrieval|> token is inserted to indicate where additional information about the transfer might be needed.",
"Ideal Knowledge": "Information about gorilla transfers between zoos, specifically detailing the move of Timmy, Tunuka, and Paki from the Bronx Zoo to the Louisville Zoo in May 2004, including reasons for the transfer and any related conservation efforts.",
"Key Factual Information": "the Louisville Zoo in May 2004",
"Prefix reformulated": "Where did Timmy and two female gorillas, Tunuka and Paki, move to from the Bronx Zoo?",
"Prefix-intent": "Timmy , along with two female gorillas , Tunuka and Paki , left the Bronx Zoo for",
"Verification": "The KFI directly provides the destination and date of the gorillas' move, as prompted by the PI."
}
]
} |
Pattycake tried her hand at painting while participating with the Wildlife Conservation Society ( WCS ) animal enrichment program at the Bronx Zoo . A sample of her work was published in the 2010 WCS Annual Report .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Pattycake tried her hand at painting while participating with the Wildlife Conservation Society ( WCS ) animal enrichment program at the Bronx Zoo . A sample of her work was published in the 2010 WCS Annual Report . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Pattycake tried her hand at painting while participating with the Wildlife Conservation Society ( WCS ) animal enrichment program at<|end-PI|> <|retrieval|> <|start-KFI|>the Bronx Zoo<|end-KFI|>. <|start-PI|>A sample of her work was published in<|end-PI|> <|retrieval|> <|start-KFI|>the 2010 WCS Annual Report<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces Pattycake's participation in the WCS program, while the KFI provides the specific location. The <|retrieval|> token is inserted to indicate where additional information about the program's location might be needed.",
"Ideal Knowledge": "Information about the Wildlife Conservation Society's animal enrichment programs, specifically mentioning the program at the Bronx Zoo and its participants, including animals like Pattycake.",
"Key Factual Information": "the Bronx Zoo",
"Prefix reformulated": "Where did Pattycake participate in the Wildlife Conservation Society (WCS) animal enrichment program?",
"Prefix-intent": "Pattycake tried her hand at painting while participating with the Wildlife Conservation Society ( WCS ) animal enrichment program at",
"Verification": "The KFI directly provides the location information requested in the PI."
},
{
"Explanation": "The PI sets up a query about where Pattycake's work was published, and the KFI provides the specific publication. The <|retrieval|> token is inserted to indicate where additional information about the publication might be needed.",
"Ideal Knowledge": "Details about the Wildlife Conservation Society's Annual Reports, specifically the 2010 edition, including information about featured animal art or enrichment program highlights.",
"Key Factual Information": "the 2010 WCS Annual Report",
"Prefix reformulated": "Where was a sample of Pattycake's painting work published?",
"Prefix-intent": "A sample of her work was published in",
"Verification": "The KFI directly provides the publication information requested in the PI."
}
]
} |
Pattycake suffered from chronic heart disease and arthritis as she aged . She was one of 338 captive zoo gorillas within North America when she died in her sleep at the age of 40 . According to the Wildlife Conservation Society , Pattycake exceeded the median life span of 37 years for female zoo gorillas .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Pattycake suffered from chronic heart disease and arthritis as she aged . She was one of 338 captive zoo gorillas within North America when she died in her sleep at the age of 40 . According to the Wildlife Conservation Society , Pattycake exceeded the median life span of 37 years for female zoo gorillas . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
In honor of Pattycake , the Rev. Frederick Douglass Kirkpatrick ( Brother Kirk ) joined Pete Seeger and the Sesame Street kids chorus for the song " Patty Cake Gorilla " , released on the album Pete Seeger and Brother Kirk Visit Sesame Street ( 1974 ) . A picture book called Patty Cake ( 1974 ) , featuring New York Times photographer Neal Boenzi and others , was written by Elizabeth Moody . Pearl Wolf wrote Gorilla Baby : The Story of Patty Cake ( 1974 ) , a picture book for children . Artist Susan Green published her direct , personal observations about the custody dispute ( along with her drawings ) in the book Gentle Gorilla : The Story of Patty Cake ( 1978 ) .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n In honor of Pattycake , the Rev. Frederick Douglass Kirkpatrick ( Brother Kirk ) joined Pete Seeger and the Sesame Street kids chorus for the song \" Patty Cake Gorilla \" , released on the album Pete Seeger and Brother Kirk Visit Sesame Street ( 1974 ) . A picture book called Patty Cake ( 1974 ) , featuring New York Times photographer Neal Boenzi and others , was written by Elizabeth Moody . Pearl Wolf wrote Gorilla Baby : The Story of Patty Cake ( 1974 ) , a picture book for children . Artist Susan Green published her direct , personal observations about the custody dispute ( along with her drawings ) in the book Gentle Gorilla : The Story of Patty Cake ( 1978 ) . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>In honor of Pattycake, the Rev. Frederick Douglass Kirkpatrick (Brother Kirk) joined Pete Seeger and the Sesame Street kids chorus for the song<|end-PI|> <|retrieval|> <|start-KFI|>\"Patty Cake Gorilla\", released on the album Pete Seeger and Brother Kirk Visit Sesame Street (1974)<|end-KFI|>. <|start-PI|>A picture book called Patty Cake (1974), featuring New York Times photographer Neal Boenzi and others, was written by<|end-PI|> <|retrieval|> <|start-KFI|>Elizabeth Moody<|end-KFI|>. <|start-PI|>Pearl Wolf wrote<|end-PI|> <|retrieval|> <|start-KFI|>Gorilla Baby: The Story of Patty Cake (1974), a picture book for children<|end-KFI|>. <|start-PI|>Artist Susan Green published her direct, personal observations about the custody dispute (along with her drawings) in the book<|end-PI|> <|retrieval|> <|start-KFI|>Gentle Gorilla: The Story of Patty Cake (1978)<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context for a song performed in honor of Pattycake, while the KFI provides the specific song title and album details. The <|retrieval|> token is inserted to indicate where additional information about the song and album might be needed.",
"Ideal Knowledge": "A detailed list of songs and albums related to Pattycake the gorilla, including collaborations between Pete Seeger, Brother Kirk, and Sesame Street, with specific focus on the 1974 album 'Pete Seeger and Brother Kirk Visit Sesame Street'.",
"Key Factual Information": "\"Patty Cake Gorilla\", released on the album Pete Seeger and Brother Kirk Visit Sesame Street (1974)",
"Prefix reformulated": "What song did Rev. Frederick Douglass Kirkpatrick, Pete Seeger, and the Sesame Street kids chorus perform in honor of Pattycake?",
"Prefix-intent": "In honor of Pattycake, the Rev. Frederick Douglass Kirkpatrick (Brother Kirk) joined Pete Seeger and the Sesame Street kids chorus for the song",
"Verification": "The KFI directly provides the song title and album information requested in the PI."
},
{
"Explanation": "The PI introduces a picture book about Patty Cake and sets up a query about its author, while the KFI provides the specific name. The <|retrieval|> token is inserted to indicate where additional information about the author might be needed.",
"Ideal Knowledge": "Biographical information about Elizabeth Moody, focusing on her work as an author of children's books, particularly those related to Pattycake the gorilla. Details about the 1974 picture book 'Patty Cake' and its collaboration with photographer Neal Boenzi.",
"Key Factual Information": "Elizabeth Moody",
"Prefix reformulated": "Who wrote the picture book called Patty Cake (1974) featuring New York Times photographer Neal Boenzi?",
"Prefix-intent": "A picture book called Patty Cake (1974), featuring New York Times photographer Neal Boenzi and others, was written by",
"Verification": "The KFI directly provides the author's name as requested in the PI."
},
{
"Explanation": "The PI introduces Pearl Wolf as an author, and the KFI provides the specific book title and type. The <|retrieval|> token is inserted to indicate where additional information about the book might be needed.",
"Ideal Knowledge": "Information about Pearl Wolf's literary works, specifically focusing on her 1974 children's picture book 'Gorilla Baby: The Story of Patty Cake'. Details about the book's content, reception, and its place in the literature about Pattycake the gorilla.",
"Key Factual Information": "Gorilla Baby: The Story of Patty Cake (1974), a picture book for children",
"Prefix reformulated": "What book did Pearl Wolf write about Pattycake?",
"Prefix-intent": "Pearl Wolf wrote",
"Verification": "The KFI directly provides the book title and type as implied by the PI."
},
{
"Explanation": "The PI introduces Susan Green's book about the custody dispute, and the KFI provides the specific book title and publication year. The <|retrieval|> token is inserted to indicate where additional information about the book and the custody dispute might be needed.",
"Ideal Knowledge": "Detailed information about Susan Green's book 'Gentle Gorilla: The Story of Patty Cake' (1978), including its content, the custody dispute it covers, and Green's personal observations and drawings. Context about the Pattycake custody case and its significance in the gorilla's life story.",
"Key Factual Information": "Gentle Gorilla: The Story of Patty Cake (1978)",
"Prefix reformulated": "What book did artist Susan Green publish about the Pattycake custody dispute?",
"Prefix-intent": "Artist Susan Green published her direct, personal observations about the custody dispute (along with her drawings) in the book",
"Verification": "The KFI directly provides the book title as requested in the PI."
}
]
} |
Lactarius indigo , commonly known as the indigo milk cap , the indigo ( or blue ) lactarius , or the blue milk mushroom , is a species of agaric fungus in the family Russulaceae . A widely distributed species , it grows naturally in eastern North America , East Asia , and Central America ; it has also been reported in southern France . L. indigo grows on the ground in both deciduous and coniferous forests , where it forms mycorrhizal associations with a broad range of trees . The fruit body color ranges from dark blue in fresh specimens to pale blue @-@ gray in older ones . The milk , or latex , that oozes when the mushroom tissue is cut or broken — a feature common to all members of the Lactarius genus — is also indigo blue , but slowly turns green upon exposure to air . The cap has a diameter of 5 to 15 cm ( 2 to 6 in ) , and the stem is 2 to 8 cm ( 0 @.@ 8 to 3 in ) tall and 1 to 2 @.@ 5 cm ( 0 @.@ 4 to 1 @.@ 0 in ) thick . It is an edible mushroom , and is sold in rural markets in China , Guatemala , and Mexico .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Lactarius indigo , commonly known as the indigo milk cap , the indigo ( or blue ) lactarius , or the blue milk mushroom , is a species of agaric fungus in the family Russulaceae . A widely distributed species , it grows naturally in eastern North America , East Asia , and Central America ; it has also been reported in southern France . L. indigo grows on the ground in both deciduous and coniferous forests , where it forms mycorrhizal associations with a broad range of trees . The fruit body color ranges from dark blue in fresh specimens to pale blue @-@ gray in older ones . The milk , or latex , that oozes when the mushroom tissue is cut or broken — a feature common to all members of the Lactarius genus — is also indigo blue , but slowly turns green upon exposure to air . The cap has a diameter of 5 to 15 cm ( 2 to 6 in ) , and the stem is 2 to 8 cm ( 0 @.@ 8 to 3 in ) tall and 1 to 2 @.@ 5 cm ( 0 @.@ 4 to 1 @.@ 0 in ) thick . It is an edible mushroom , and is sold in rural markets in China , Guatemala , and Mexico . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "<|start-PI|>Lactarius indigo, commonly known as the indigo milk cap, the indigo (or blue) lactarius, or the blue milk mushroom, is<|end-PI|> <|retrieval|> <|start-KFI|>a species of agaric fungus in the family Russulaceae<|end-KFI|>. <|start-PI|>A widely distributed species, it grows naturally in<|end-PI|> <|retrieval|> <|start-KFI|>eastern North America, East Asia, and Central America; it has also been reported in southern France<|end-KFI|>. <|start-PI|>L. indigo grows on the ground in both deciduous and coniferous forests, where it forms<|end-PI|> <|retrieval|> <|start-KFI|>mycorrhizal associations with a broad range of trees<|end-KFI|>. <|start-PI|>The fruit body color ranges from<|end-PI|> <|retrieval|> <|start-KFI|>dark blue in fresh specimens to pale blue-gray in older ones<|end-KFI|>. <|start-PI|>The milk, or latex, that oozes when the mushroom tissue is cut or broken — a feature common to all members of the Lactarius genus — is also indigo blue, but<|end-PI|> <|retrieval|> <|start-KFI|>slowly turns green upon exposure to air<|end-KFI|>. <|start-PI|>The cap has a diameter of<|end-PI|> <|retrieval|> <|start-KFI|>5 to 15 cm (2 to 6 in)<|end-KFI|>, <|start-PI|>and the stem is<|end-PI|> <|retrieval|> <|start-KFI|>2 to 8 cm (0.8 to 3 in) tall and 1 to 2.5 cm (0.4 to 1.0 in) thick<|end-KFI|>. <|start-PI|>It is an edible mushroom, and is sold in<|end-PI|> <|retrieval|> <|start-KFI|>rural markets in China, Guatemala, and Mexico<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces Lactarius indigo and its common names, setting up a query about its classification. The KFI provides the specific taxonomic information. The <|retrieval|> token is inserted to indicate where additional classification details might be needed.",
"Ideal Knowledge": "A detailed taxonomic classification of Lactarius indigo, including its kingdom, phylum, class, order, family, genus, and species, with emphasis on its placement within the Russulaceae family.",
"Key Factual Information": "a species of agaric fungus in the family Russulaceae",
"Prefix reformulated": "What type of organism is Lactarius indigo?",
"Prefix-intent": "Lactarius indigo, commonly known as the indigo milk cap, the indigo (or blue) lactarius, or the blue milk mushroom, is",
"Verification": "The KFI directly provides the classification of Lactarius indigo as requested in the PI."
},
{
"Explanation": "The PI sets up a query about the natural distribution of the species, and the KFI provides specific geographical locations. The <|retrieval|> token is inserted to indicate where additional distribution information might be needed.",
"Ideal Knowledge": "A comprehensive list of regions and countries where Lactarius indigo is found, including specific ecological zones and habitat types within these areas.",
"Key Factual Information": "eastern North America, East Asia, and Central America; it has also been reported in southern France",
"Prefix reformulated": "Where does Lactarius indigo grow naturally?",
"Prefix-intent": "A widely distributed species, it grows naturally in",
"Verification": "The KFI directly lists the geographical locations where Lactarius indigo naturally grows, as prompted by the PI."
},
{
"Explanation": "The PI describes the habitat of L. indigo and sets up a query about its ecological relationships. The KFI provides information about its mycorrhizal associations. The <|retrieval|> token is inserted to indicate where additional ecological information might be needed.",
"Ideal Knowledge": "Detailed information on mycorrhizal associations, including the specific tree species Lactarius indigo associates with, and the ecological benefits of these relationships.",
"Key Factual Information": "mycorrhizal associations with a broad range of trees",
"Prefix reformulated": "What type of associations does Lactarius indigo form in forests?",
"Prefix-intent": "L. indigo grows on the ground in both deciduous and coniferous forests, where it forms",
"Verification": "The KFI directly describes the type of associations Lactarius indigo forms in forests, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the color of the fruit body, and the KFI provides the specific color range and how it changes with age. The <|retrieval|> token is inserted to indicate where additional information about color variations might be needed.",
"Ideal Knowledge": "Detailed description of the color changes in Lactarius indigo fruit bodies over time, including the chemical processes responsible for these changes and how to identify the mushroom at different stages of maturity.",
"Key Factual Information": "dark blue in fresh specimens to pale blue-gray in older ones",
"Prefix reformulated": "What is the color range of Lactarius indigo's fruit body?",
"Prefix-intent": "The fruit body color ranges from",
"Verification": "The KFI directly describes the color range of the fruit body, as prompted by the PI."
},
{
"Explanation": "The PI describes the initial color of the mushroom's milk and sets up a query about what happens to it. The KFI provides information about the color change when exposed to air. The <|retrieval|> token is inserted to indicate where additional information about this chemical process might be needed.",
"Ideal Knowledge": "Detailed explanation of the chemical processes causing the color change in Lactarius indigo's milk when exposed to air, including the specific compounds involved and the time frame of the color change.",
"Key Factual Information": "slowly turns green upon exposure to air",
"Prefix reformulated": "What happens to the color of Lactarius indigo's milk when exposed to air?",
"Prefix-intent": "The milk, or latex, that oozes when the mushroom tissue is cut or broken — a feature common to all members of the Lactarius genus — is also indigo blue, but",
"Verification": "The KFI directly describes the color change of the mushroom's milk when exposed to air, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the size of the mushroom cap, and the KFI provides the specific size range. The <|retrieval|> token is inserted to indicate where additional information about cap size variations might be needed.",
"Ideal Knowledge": "Detailed information on the size variations of Lactarius indigo caps, including factors that influence size such as age, growing conditions, and genetic variations.",
"Key Factual Information": "5 to 15 cm (2 to 6 in)",
"Prefix reformulated": "What is the diameter of Lactarius indigo's cap?",
"Prefix-intent": "The cap has a diameter of",
"Verification": "The KFI directly provides the size range of the mushroom cap, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the stem dimensions, and the KFI provides specific height and thickness ranges. The <|retrieval|> token is inserted to indicate where additional information about stem morphology might be needed.",
"Ideal Knowledge": "Comprehensive information on Lactarius indigo stem morphology, including factors affecting stem dimensions, typical growth patterns, and how stem characteristics can be used for identification.",
"Key Factual Information": "2 to 8 cm (0.8 to 3 in) tall and 1 to 2.5 cm (0.4 to 1.0 in) thick",
"Prefix reformulated": "What are the dimensions of Lactarius indigo's stem?",
"Prefix-intent": "and the stem is",
"Verification": "The KFI directly provides the height and thickness ranges of the mushroom stem, as prompted by the PI."
},
{
"Explanation": "The PI establishes that the mushroom is edible and sold, and the KFI provides specific locations where it is marketed. The <|retrieval|> token is inserted to indicate where additional information about its culinary use and market presence might be needed.",
"Ideal Knowledge": "Detailed information on the culinary use and market presence of Lactarius indigo in various countries, including traditional recipes, preparation methods, and cultural significance in each region.",
"Key Factual Information": "rural markets in China, Guatemala, and Mexico",
"Prefix reformulated": "Where is Lactarius indigo sold as an edible mushroom?",
"Prefix-intent": "It is an edible mushroom, and is sold in",
"Verification": "The KFI directly lists the locations where the mushroom is sold in markets, as prompted by the PI."
}
]
} |
Originally described in 1822 as Agaricus indigo by American mycologist Lewis David de Schweinitz , the species was later transferred to the genus Lactarius in 1838 by the Swede Elias Magnus Fries . German botanist Otto Kuntze called it Lactifluus indigo in his 1891 treatise Revisio Generum Plantarum , but the suggested name change was not adopted by others . Hesler and Smith in their 1960 study of North American species of Lactarius defined L. indigo as the type species of subsection Caerulei , a group characterized by blue latex and a sticky , blue cap . In 1979 , they revised their opinions on the organization of subdivisions in the genus Lactarius , and instead placed L. indigo in subgenus Lactarius based on the color of latex , and the subsequent color changes observed after exposure to air . As they explained :
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Originally described in 1822 as Agaricus indigo by American mycologist Lewis David de Schweinitz , the species was later transferred to the genus Lactarius in 1838 by the Swede Elias Magnus Fries . German botanist Otto Kuntze called it Lactifluus indigo in his 1891 treatise Revisio Generum Plantarum , but the suggested name change was not adopted by others . Hesler and Smith in their 1960 study of North American species of Lactarius defined L. indigo as the type species of subsection Caerulei , a group characterized by blue latex and a sticky , blue cap . In 1979 , they revised their opinions on the organization of subdivisions in the genus Lactarius , and instead placed L. indigo in subgenus Lactarius based on the color of latex , and the subsequent color changes observed after exposure to air . As they explained : \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
The gradual development of blue to violet pigmentation as one progresses from species to species is an interesting phenomenon deserving further study . The climax is reached in L. indigo which is blue throughout . L. chelidonium and its variety chelidonioides , L. paradoxus , and L. hemicyaneus may be considered as mileposts along the road to L. indigo .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n The gradual development of blue to violet pigmentation as one progresses from species to species is an interesting phenomenon deserving further study . The climax is reached in L. indigo which is blue throughout . L. chelidonium and its variety chelidonioides , L. paradoxus , and L. hemicyaneus may be considered as mileposts along the road to L. indigo . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
The specific epithet indigo is derived from the Latin word meaning " indigo blue " . Its names in the English vernacular include the " indigo milk cap " , the " indigo Lactarius " , the " blue milk mushroom " , and the " blue Lactarius " . In central Mexico , it is known as añil , azul , hongo azul , zuin , and zuine ; it is also called quexque ( meaning " blue " ) in Veracruz and Puebla .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n The specific epithet indigo is derived from the Latin word meaning \" indigo blue \" . Its names in the English vernacular include the \" indigo milk cap \" , the \" indigo Lactarius \" , the \" blue milk mushroom \" , and the \" blue Lactarius \" . In central Mexico , it is known as añil , azul , hongo azul , zuin , and zuine ; it is also called quexque ( meaning \" blue \" ) in Veracruz and Puebla . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
Like many other mushrooms , L. indigo develops from a nodule , or pinhead , that forms within the underground mycelium , a mass of threadlike fungal cells called hyphae that make up the bulk of the organism . Under appropriate environmental conditions of temperature , humidity , and nutrient availability , the visible reproductive structures ( fruit bodies ) are formed . The cap of the fruit body , measuring between 5 and 15 cm ( 2 @.@ 0 and 5 @.@ 9 in ) in diameter , is initially convex and later develops a central depression ; in age it becomes even more deeply depressed , becoming somewhat funnel @-@ shaped as the edge of the cap lifts upward . The margin of the cap is rolled inwards when young , but unrolls and elevates as it matures . The cap surface is indigo blue when fresh , but fades to a paler grayish- or silvery @-@ blue , sometimes with greenish splotches . It is often zonate : marked with concentric lines that form alternating pale and darker zones , and the cap may have dark blue spots , especially towards the edge . Young caps are sticky to the touch .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Like many other mushrooms , L. indigo develops from a nodule , or pinhead , that forms within the underground mycelium , a mass of threadlike fungal cells called hyphae that make up the bulk of the organism . Under appropriate environmental conditions of temperature , humidity , and nutrient availability , the visible reproductive structures ( fruit bodies ) are formed . The cap of the fruit body , measuring between 5 and 15 cm ( 2 @.@ 0 and 5 @.@ 9 in ) in diameter , is initially convex and later develops a central depression ; in age it becomes even more deeply depressed , becoming somewhat funnel @-@ shaped as the edge of the cap lifts upward . The margin of the cap is rolled inwards when young , but unrolls and elevates as it matures . The cap surface is indigo blue when fresh , but fades to a paler grayish- or silvery @-@ blue , sometimes with greenish splotches . It is often zonate : marked with concentric lines that form alternating pale and darker zones , and the cap may have dark blue spots , especially towards the edge . Young caps are sticky to the touch . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "Like many other mushrooms, <|start-PI|> L. indigo develops from a nodule, or pinhead, that forms within <|end-PI|> <|retrieval|> <|start-KFI|> the underground mycelium, a mass of threadlike fungal cells called hyphae that make up the bulk of the organism <|end-KFI|>. Under appropriate environmental conditions of temperature, humidity, and nutrient availability, the visible reproductive structures (fruit bodies) are formed. <|start-PI|> The cap of the fruit body, measuring between <|end-PI|> <|retrieval|> <|start-KFI|> 5 and 15 cm (2.0 and 5.9 in) in diameter <|end-KFI|>, is initially convex and later develops a central depression; in age it becomes even more deeply depressed, becoming somewhat funnel-shaped as the edge of the cap lifts upward. The margin of the cap is rolled inwards when young, but unrolls and elevates as it matures. <|start-PI|> The cap surface is indigo blue when fresh, but fades to <|end-PI|> <|retrieval|> <|start-KFI|> a paler grayish- or silvery-blue, sometimes with greenish splotches <|end-KFI|>. It is often zonate: marked with concentric lines that form alternating pale and darker zones, and the cap may have dark blue spots, especially towards the edge. Young caps are sticky to the touch.",
"none": null,
"results": [
{
"Explanation": "The PI introduces L. indigo and sets up a query about its development, while the KFI provides specific information about its origin. The <|retrieval|> token is inserted to indicate where additional information about mushroom development might be needed.",
"Ideal Knowledge": "Detailed information about the life cycle of L. indigo mushrooms, focusing on the early stages of development and the role of mycelium in fungal growth.",
"Key Factual Information": "the underground mycelium, a mass of threadlike fungal cells called hyphae that make up the bulk of the organism",
"Prefix reformulated": "Where does L. indigo mushroom develop from?",
"Prefix-intent": "Like many other mushrooms, L. indigo develops from a nodule, or pinhead, that forms within",
"Verification": "The KFI directly provides information about where L. indigo develops, as prompted by the PI."
},
{
"Explanation": "The PI introduces the cap of the fruit body and sets up a query about its size, while the KFI provides specific measurements. The <|retrieval|> token is inserted to indicate where additional information about mushroom cap sizes might be needed.",
"Ideal Knowledge": "A comprehensive guide to L. indigo mushroom characteristics, including detailed measurements of various parts of the fruit body at different growth stages.",
"Key Factual Information": "5 and 15 cm (2.0 and 5.9 in) in diameter",
"Prefix reformulated": "What is the size range of the L. indigo mushroom cap?",
"Prefix-intent": "The cap of the fruit body, measuring between",
"Verification": "The KFI directly provides the size range of the mushroom cap, as prompted by the PI."
},
{
"Explanation": "The PI describes the initial color of the cap and sets up a query about how it changes, while the KFI provides specific information about the faded colors. The <|retrieval|> token is inserted to indicate where additional information about mushroom color changes might be needed.",
"Ideal Knowledge": "Detailed information about the color changes in L. indigo mushrooms over time, including the chemical processes responsible for these changes and factors that influence color intensity.",
"Key Factual Information": "a paler grayish- or silvery-blue, sometimes with greenish splotches",
"Prefix reformulated": "What color does the L. indigo mushroom cap fade to?",
"Prefix-intent": "The cap surface is indigo blue when fresh, but fades to",
"Verification": "The KFI directly describes the color changes of the mushroom cap, as prompted by the PI."
}
]
} |
The flesh is pallid to bluish in color , slowly turning greenish after being exposed to air ; its taste is mild to slightly acrid . The flesh of the entire mushroom is brittle , and the stem , if bent sufficiently , will snap open cleanly . The latex exuded from injured tissue is indigo blue , and stains the wounded tissue greenish ; like the flesh , the latex has a mild taste . Lactarius indigo is noted for not producing as much latex as other Lactarius species , and older specimens in particular may be too dried out to produce any latex .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n The flesh is pallid to bluish in color , slowly turning greenish after being exposed to air ; its taste is mild to slightly acrid . The flesh of the entire mushroom is brittle , and the stem , if bent sufficiently , will snap open cleanly . The latex exuded from injured tissue is indigo blue , and stains the wounded tissue greenish ; like the flesh , the latex has a mild taste . Lactarius indigo is noted for not producing as much latex as other Lactarius species , and older specimens in particular may be too dried out to produce any latex . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
The gills of the mushroom range from adnate ( squarely attached to the stem ) to slightly decurrent ( running down the length of the stem ) , and crowded close together . Their color is an indigo blue , becoming paler with age or staining green with damage . The stem is 2 – 6 cm ( 0 @.@ 8 – 2 @.@ 4 in ) tall by 1 – 2 @.@ 5 cm ( 0 @.@ 4 – 1 @.@ 0 in ) thick , and the same diameter throughout or sometimes narrowed at base . Its color is indigo blue to silvery- or grayish blue . The interior of the stem is solid and firm initially , but develops a hollow with age . Like the cap , it is initially sticky or slimy to the touch when young , but soon dries out . Its attachment to the cap is usually in a central position , although it may also be off @-@ center . Fruit bodies of L. indigo have no distinguishable odor .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n The gills of the mushroom range from adnate ( squarely attached to the stem ) to slightly decurrent ( running down the length of the stem ) , and crowded close together . Their color is an indigo blue , becoming paler with age or staining green with damage . The stem is 2 – 6 cm ( 0 @.@ 8 – 2 @.@ 4 in ) tall by 1 – 2 @.@ 5 cm ( 0 @.@ 4 – 1 @.@ 0 in ) thick , and the same diameter throughout or sometimes narrowed at base . Its color is indigo blue to silvery- or grayish blue . The interior of the stem is solid and firm initially , but develops a hollow with age . Like the cap , it is initially sticky or slimy to the touch when young , but soon dries out . Its attachment to the cap is usually in a central position , although it may also be off @-@ center . Fruit bodies of L. indigo have no distinguishable odor . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |