text
stringlengths 101
3.69k
| messages
listlengths 1
1
| outputs
dict |
|---|---|---|
The barracks consists of two levels , with rifle firing holes on all sides . Iron shutters closed on the inside and were locked with wooden beams . In 1885 the verandah facing the manning parade was enclosed with wooden shutters to keep the weather out . These shutters were removed during 20th @-@ century restoration work and the verandah returned to original condition . The roof was at first flat timber overlaid by 12 inches ( 30 cm ) of lime concrete . This first roof leaked and an iron roof was added in 1885 . The first floor of the barracks contains the officer 's rooms and troop 's barracks room . Its eastern ( outer ) wall is not flat , incorporating a design feature known as " hornwork " , which opens the field of fire from the rifle holes . Some of the firing holes were bricked in during the 1930s ; this has been only partly rectified during restoration .
| [
{
"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 barracks consists of two levels , with rifle firing holes on all sides . Iron shutters closed on the inside and were locked with wooden beams . In 1885 the verandah facing the manning parade was enclosed with wooden shutters to keep the weather out . These shutters were removed during 20th @-@ century restoration work and the verandah returned to original condition . The roof was at first flat timber overlaid by 12 inches ( 30 cm ) of lime concrete . This first roof leaked and an iron roof was added in 1885 . The first floor of the barracks contains the officer 's rooms and troop 's barracks room . Its eastern ( outer ) wall is not flat , incorporating a design feature known as \" hornwork \" , which opens the field of fire from the rifle holes . Some of the firing holes were bricked in during the 1930s ; this has been only partly rectified during restoration . \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 barracks consists of two levels , with<|end-PI|> <|retrieval|> <|start-KFI|>rifle firing holes on all sides<|end-KFI|> . Iron shutters closed on the inside and were locked with wooden beams . <|start-PI|>In 1885 the verandah facing the manning parade was enclosed with<|end-PI|> <|retrieval|> <|start-KFI|>wooden shutters to keep the weather out<|end-KFI|> . These shutters were removed during 20th @-@ century restoration work and the verandah returned to original condition . <|start-PI|>The roof was at first flat timber overlaid by<|end-PI|> <|retrieval|> <|start-KFI|>12 inches ( 30 cm ) of lime concrete<|end-KFI|> . This first roof leaked and an iron roof was added in 1885 . <|start-PI|>The first floor of the barracks contains<|end-PI|> <|retrieval|> <|start-KFI|>the officer 's rooms and troop 's barracks room<|end-KFI|> . <|start-PI|>Its eastern ( outer ) wall is not flat , incorporating a design feature known as<|end-PI|> <|retrieval|> <|start-KFI|>\" hornwork \" , which opens the field of fire from the rifle holes<|end-KFI|> . Some of the firing holes were bricked in during the 1930s ; this has been only partly rectified during restoration .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the barracks and sets up a query about its features, while the KFI provides specific information about the rifle firing holes. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "Detailed description of military barracks architecture, specifically focusing on defensive features such as rifle firing holes and their placement.",
"Key Factual Information": "rifle firing holes on all sides",
"Prefix reformulated": "What features do the barracks have?",
"Prefix-intent": "The barracks consists of two levels , with",
"Verification": "The KFI directly provides information about the features of the barracks as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about a specific modification made to the barracks in 1885, and the KFI provides the details of this modification. The <|retrieval|> token is inserted to indicate where additional historical information might be needed.",
"Ideal Knowledge": "Historical records of modifications made to military barracks in 1885, specifically detailing the addition of wooden shutters to verandahs for weather protection.",
"Key Factual Information": "wooden shutters to keep the weather out",
"Prefix reformulated": "How was the verandah facing the manning parade enclosed in 1885?",
"Prefix-intent": "In 1885 the verandah facing the manning parade was enclosed with",
"Verification": "The KFI directly provides information about how the verandah was enclosed, as prompted by the PI."
},
{
"Explanation": "The PI introduces the original roof structure, and the KFI provides specific measurements and materials. The <|retrieval|> token is inserted to indicate where additional information about historical construction techniques might be needed.",
"Ideal Knowledge": "Detailed information about historical roofing techniques, specifically the use of flat timber overlaid with lime concrete in military structures.",
"Key Factual Information": "12 inches ( 30 cm ) of lime concrete",
"Prefix reformulated": "What was the original roof of the barracks made of?",
"Prefix-intent": "The roof was at first flat timber overlaid by",
"Verification": "The KFI directly provides information about the original roof composition, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the contents of the first floor, and the KFI provides specific information about the rooms. The <|retrieval|> token is inserted to indicate where additional information about barracks layout might be needed.",
"Ideal Knowledge": "Layout plans of historical military barracks, detailing the typical arrangement of rooms and their purposes on different floors.",
"Key Factual Information": "the officer 's rooms and troop 's barracks room",
"Prefix reformulated": "What does the first floor of the barracks contain?",
"Prefix-intent": "The first floor of the barracks contains",
"Verification": "The KFI directly provides information about the contents of the first floor, as prompted by the PI."
},
{
"Explanation": "The PI introduces a unique design feature of the eastern wall, and the KFI provides the name and purpose of this feature. The <|retrieval|> token is inserted to indicate where additional information about military architectural features might be needed.",
"Ideal Knowledge": "Detailed explanations of military architecture, specifically focusing on defensive design features like 'hornwork' and their strategic purposes.",
"Key Factual Information": "\" hornwork \" , which opens the field of fire from the rifle holes",
"Prefix reformulated": "What design feature is incorporated into the eastern wall of the barracks?",
"Prefix-intent": "Its eastern ( outer ) wall is not flat , incorporating a design feature known as",
"Verification": "The KFI directly provides information about the specific design feature and its purpose, as prompted by the PI."
}
]
} |
The basement contains the gunner 's mess , canteen , No. 1 ancillary store , pantry and officer 's kitchen . The canteen sold everyday necessities as well as a few items like biscuits , sweets and tobacco to the stationed troops . Profits from the canteen were used to fund sporting equipment for the garrison . The No.1 ancillary store was used for various pieces of delicate equipment , fuses , friction tubes and rockets . In 1887 an explosion in the room caused a number of injuries , damage to the store and to the above officer 's quarters . The barracks room and officer 's quarters are connected with a door , probably added in the 1930s as it is not part of the original plans . The barracks room accommodated approximately 20 men . For display purposes it is outfitted as for similar period barracks .
| [
{
"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 basement contains the gunner 's mess , canteen , No. 1 ancillary store , pantry and officer 's kitchen . The canteen sold everyday necessities as well as a few items like biscuits , sweets and tobacco to the stationed troops . Profits from the canteen were used to fund sporting equipment for the garrison . The No.1 ancillary store was used for various pieces of delicate equipment , fuses , friction tubes and rockets . In 1887 an explosion in the room caused a number of injuries , damage to the store and to the above officer 's quarters . The barracks room and officer 's quarters are connected with a door , probably added in the 1930s as it is not part of the original plans . The barracks room accommodated approximately 20 men . For display purposes it is outfitted as for similar period barracks . \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 laboratory is built into the rear wall , on the forts southern side . This room is set into the end of the rampart and was used to prepare gunpowder charges . A guardhouse was built at the manning parade 's southern end in 1885 . It is now used as a duty room for the drill squad during recreation demonstrations but originally was a guard 's room and separate cells connected by telephone to Fort Largs . Up to three men appear to have been accommodated in the guard 's room . When the fort was converted to use as a caravan park the guard house was converted to an ablution block . A store and ablution block were also built in 1885 between the stockade and rear defence wall .
| [
{
"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 laboratory is built into the rear wall , on the forts southern side . This room is set into the end of the rampart and was used to prepare gunpowder charges . A guardhouse was built at the manning parade 's southern end in 1885 . It is now used as a duty room for the drill squad during recreation demonstrations but originally was a guard 's room and separate cells connected by telephone to Fort Largs . Up to three men appear to have been accommodated in the guard 's room . When the fort was converted to use as a caravan park the guard house was converted to an ablution block . A store and ablution block were also built in 1885 between the stockade and rear defence wall . \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
} |
Early plans for the fort 's armament were drawn up by Harding Stewart of the British War Office . They called for four 9 in ( 230 mm ) 12 long tons ( 12 t ) rifled muzzle @-@ loading ( RML ) guns , two mounted in turrets and two behind vertical iron shields . This configuration of siege artillery had not been tried before and the plan was abandoned due to the large expected cost . Two of the 9 in guns had been in South Australia since c.1868 after a Colonial Government request , but were not used in the eventual fort design . The final Jervois / Scratchley design omitted the turret and iron shields . 64 @-@ pounder RML guns were substituted for two of the 9 in guns and 10 in 20 ton RML guns for the remaining two . The 20 ton guns were chosen over the then standard 18 ton gun by Jervois after he saw the plans in England , the decision influenced by the then fully committed nature of the Royal Arsenal . Jervois had originally ordered the 18 ton guns but cancelled the order and changed to the Armstrong 20 ton when the arsenal was unable fulfil the order . He also ordered Armstrong 's mechanical loading and protected barbette system for the 20 ton guns . This original battery of four heavy guns remains in place at the fort .
| [
{
"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 plans for the fort 's armament were drawn up by Harding Stewart of the British War Office . They called for four 9 in ( 230 mm ) 12 long tons ( 12 t ) rifled muzzle @-@ loading ( RML ) guns , two mounted in turrets and two behind vertical iron shields . This configuration of siege artillery had not been tried before and the plan was abandoned due to the large expected cost . Two of the 9 in guns had been in South Australia since c.1868 after a Colonial Government request , but were not used in the eventual fort design . The final Jervois / Scratchley design omitted the turret and iron shields . 64 @-@ pounder RML guns were substituted for two of the 9 in guns and 10 in 20 ton RML guns for the remaining two . The 20 ton guns were chosen over the then standard 18 ton gun by Jervois after he saw the plans in England , the decision influenced by the then fully committed nature of the Royal Arsenal . Jervois had originally ordered the 18 ton guns but cancelled the order and changed to the Armstrong 20 ton when the arsenal was unable fulfil the order . He also ordered Armstrong 's mechanical loading and protected barbette system for the 20 ton guns . This original battery of four heavy guns remains in place at the fort . \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 plans for the fort 's armament were drawn up by <|start-PI|> Early plans for the fort's armament were drawn up by <|end-PI|> <|retrieval|> <|start-KFI|> Harding Stewart of the British War Office <|end-KFI|> . <|start-PI|> They called for four 9 in ( 230 mm ) 12 long tons ( 12 t ) rifled muzzle @-@ loading ( RML ) guns , <|end-PI|> <|retrieval|> <|start-KFI|> two mounted in turrets and two behind vertical iron shields <|end-KFI|> . This configuration of siege artillery had not been tried before and the plan was abandoned due to the large expected cost . <|start-PI|> Two of the 9 in guns had been in South Australia since <|end-PI|> <|retrieval|> <|start-KFI|> c.1868 after a Colonial Government request <|end-KFI|> , but were not used in the eventual fort design . <|start-PI|> The final Jervois / Scratchley design omitted the turret and iron shields . <|end-PI|> <|retrieval|> <|start-KFI|> 64 @-@ pounder RML guns were substituted for two of the 9 in guns and 10 in 20 ton RML guns for the remaining two <|end-KFI|> . The 20 ton guns were chosen over the then standard 18 ton gun by Jervois after he saw the plans in England , the decision influenced by the then fully committed nature of the Royal Arsenal . <|start-PI|> Jervois had originally ordered the 18 ton guns but cancelled the order and changed to <|end-PI|> <|retrieval|> <|start-KFI|> the Armstrong 20 ton when the arsenal was unable fulfil the order <|end-KFI|> . He also ordered Armstrong 's mechanical loading and protected barbette system for the 20 ton guns . This original battery of four heavy guns remains in place at the fort .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the topic of early plans for the fort's armament, while the KFI provides the specific person responsible. The <|retrieval|> token is inserted to indicate where additional information about Harding Stewart might be needed.",
"Ideal Knowledge": "Biographical information about Harding Stewart, his role in the British War Office, and his involvement in fort armament planning during the relevant time period.",
"Key Factual Information": "Harding Stewart of the British War Office",
"Prefix reformulated": "Who drew up the early plans for the fort's armament?",
"Prefix-intent": "Early plans for the fort's armament were drawn up by",
"Verification": "The KFI directly provides the name and affiliation of the person who drew up the plans, as requested in the PI."
},
{
"Explanation": "The PI describes the type and number of guns, while the KFI provides their specific placement. The <|retrieval|> token is inserted to indicate where additional information about gun configurations might be needed.",
"Ideal Knowledge": "Detailed information about 19th-century fort armament configurations, including the use of turrets and vertical iron shields for mounting heavy guns.",
"Key Factual Information": "two mounted in turrets and two behind vertical iron shields",
"Prefix reformulated": "What was the configuration of guns called for in the early plans for the fort's armament?",
"Prefix-intent": "They called for four 9 in ( 230 mm ) 12 long tons ( 12 t ) rifled muzzle @-@ loading ( RML ) guns ,",
"Verification": "The KFI directly provides the specific configuration of the guns mentioned in the PI."
},
{
"Explanation": "The PI sets up a query about the presence of specific guns in South Australia, and the KFI provides the timeframe and context. The <|retrieval|> token is inserted to indicate where additional historical information might be needed.",
"Ideal Knowledge": "Historical records of artillery shipments to South Australia, including dates, reasons for requests, and details about the Colonial Government's military preparations during that period.",
"Key Factual Information": "c.1868 after a Colonial Government request",
"Prefix reformulated": "When did two of the 9-inch guns arrive in South Australia?",
"Prefix-intent": "Two of the 9 in guns had been in South Australia since",
"Verification": "The KFI directly provides the approximate date and reason for the guns' presence in South Australia, as prompted by the PI."
},
{
"Explanation": "The PI introduces the final design changes, while the KFI provides the specific alterations to the gun types. The <|retrieval|> token is inserted to indicate where additional information about these gun types and design decisions might be needed.",
"Ideal Knowledge": "Detailed specifications of 64-pounder RML guns and 10-inch 20-ton RML guns, including their advantages over the originally planned 9-inch guns, and reasons for the design changes made by Jervois and Scratchley.",
"Key Factual Information": "64 @-@ pounder RML guns were substituted for two of the 9 in guns and 10 in 20 ton RML guns for the remaining two",
"Prefix reformulated": "What changes were made in the final Jervois / Scratchley design for the fort's armament?",
"Prefix-intent": "The final Jervois / Scratchley design omitted the turret and iron shields .",
"Verification": "The KFI directly describes the changes made to the gun configuration in the final design, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of a change in gun orders, and the KFI provides the specific new gun type and reason for the change. The <|retrieval|> token is inserted to indicate where additional information about these guns and the arsenal's capabilities might be needed.",
"Ideal Knowledge": "Information about Armstrong 20-ton guns, their specifications, advantages over 18-ton guns, and the production capabilities of the Royal Arsenal during that period.",
"Key Factual Information": "the Armstrong 20 ton when the arsenal was unable fulfil the order",
"Prefix reformulated": "What type of guns did Jervois change to after cancelling the order for 18-ton guns?",
"Prefix-intent": "Jervois had originally ordered the 18 ton guns but cancelled the order and changed to",
"Verification": "The KFI directly provides the new gun type and the reason for the change, as prompted by the PI."
}
]
} |
The fort 's main armament were the two RML 10 inch 20 ton guns — Numbers 3470 and 3472 . They were manufactured in 1879 by WG Armstrong and Company and originally used the Armstrong protected barbette loading system . They were made to the 1878 pattern and supplied on traversing slides . The guns have a 10 @-@ inch ( 250 mm ) calibre , a range of 6 @,@ 500 yards ( 5 @,@ 900 m ) with a muzzle velocity of 1 @,@ 630 feet per second ( 497 m / s ) and are capable of penetrating 11 inches ( 28 cm ) of iron at 2 @,@ 000 yards ( 1 @,@ 800 m ) .
| [
{
"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 fort 's main armament were the two RML 10 inch 20 ton guns — Numbers 3470 and 3472 . They were manufactured in 1879 by WG Armstrong and Company and originally used the Armstrong protected barbette loading system . They were made to the 1878 pattern and supplied on traversing slides . The guns have a 10 @-@ inch ( 250 mm ) calibre , a range of 6 @,@ 500 yards ( 5 @,@ 900 m ) with a muzzle velocity of 1 @,@ 630 feet per second ( 497 m / s ) and are capable of penetrating 11 inches ( 28 cm ) of iron at 2 @,@ 000 yards ( 1 @,@ 800 m ) . \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 fire 400 @-@ pound ( 181 kg ) projectiles using a 130 @-@ pound ( 59 kg ) gunpowder cartridge . Though they were insufficient against contemporary battleships , they were seen as adequate for the light cruisers that the fort was expected to face . The carriages weighed 13 long tons ( 13 t ) and moved on semi @-@ circular traverses . The guns were loaded either manually — using the Woolwich ( derrick ) system — or via the Armstrong mechanical loader . The Armstrong system enabled loading of the guns from behind the protection of the rampart whereas the manual system required at least two crew to be exposed on the rampart 's top . Despite this the guns were manually loaded for most of their operational life as the Armstrong cable drive was faulty and too difficult to maintain . The guns had a 14 @-@ man crew and loading took about 2 minutes , a rate of fire that was only slightly faster using the Armstrong equipment .
| [
{
"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 fire 400 @-@ pound ( 181 kg ) projectiles using a 130 @-@ pound ( 59 kg ) gunpowder cartridge . Though they were insufficient against contemporary battleships , they were seen as adequate for the light cruisers that the fort was expected to face . The carriages weighed 13 long tons ( 13 t ) and moved on semi @-@ circular traverses . The guns were loaded either manually — using the Woolwich ( derrick ) system — or via the Armstrong mechanical loader . The Armstrong system enabled loading of the guns from behind the protection of the rampart whereas the manual system required at least two crew to be exposed on the rampart 's top . Despite this the guns were manually loaded for most of their operational life as the Armstrong cable drive was faulty and too difficult to maintain . The guns had a 14 @-@ man crew and loading took about 2 minutes , a rate of fire that was only slightly faster using the Armstrong equipment . \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|>They fire 400 @-@ pound ( 181 kg ) projectiles using<|end-PI|> <|retrieval|> <|start-KFI|>a 130 @-@ pound ( 59 kg ) gunpowder cartridge<|end-KFI|>. Though they were insufficient against contemporary battleships , they were seen as adequate for the light cruisers that the fort was expected to face . <|start-PI|>The carriages weighed 13 long tons ( 13 t ) and moved on<|end-PI|> <|retrieval|> <|start-KFI|>semi @-@ circular traverses<|end-KFI|>. <|start-PI|>The guns were loaded either manually — using the Woolwich ( derrick ) system — or via<|end-PI|> <|retrieval|> <|start-KFI|>the Armstrong mechanical loader<|end-KFI|>. The Armstrong system enabled loading of the guns from behind the protection of the rampart whereas the manual system required at least two crew to be exposed on the rampart 's top . Despite this the guns were manually loaded for most of their operational life as the Armstrong cable drive was faulty and too difficult to maintain . <|start-PI|>The guns had a 14 @-@ man crew and loading took about<|end-PI|> <|retrieval|> <|start-KFI|>2 minutes , a rate of fire that was only slightly faster using the Armstrong equipment<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI sets up information about the projectiles fired, and the KFI provides the specific details about the gunpowder cartridge used. The <|retrieval|> token is inserted to indicate where additional information about the firing mechanism might be needed.",
"Ideal Knowledge": "Detailed specifications of historical coastal defense guns, including projectile weights and propellant charges used in various models.",
"Key Factual Information": "a 130 @-@ pound ( 59 kg ) gunpowder cartridge",
"Prefix reformulated": "What do the guns at the fort use to fire 400-pound (181 kg) projectiles?",
"Prefix-intent": "They fire 400 @-@ pound ( 181 kg ) projectiles using",
"Verification": "The KFI directly provides the information about the gunpowder cartridge used to fire the projectiles, as prompted by the PI."
},
{
"Explanation": "The PI introduces the weight of the carriages and sets up a query about their movement, while the KFI provides the specific type of traverse used. The <|retrieval|> token is inserted to indicate where additional information about gun carriage designs might be needed.",
"Ideal Knowledge": "Information on various types of gun carriage designs used in coastal fortifications, with emphasis on traverse mechanisms and their advantages.",
"Key Factual Information": "semi @-@ circular traverses",
"Prefix reformulated": "What type of mechanism did the 13-ton gun carriages move on?",
"Prefix-intent": "The carriages weighed 13 long tons ( 13 t ) and moved on",
"Verification": "The KFI directly describes the movement mechanism of the gun carriages, as prompted by the PI."
},
{
"Explanation": "The PI sets up information about the loading systems, mentioning the manual option, and the KFI provides the specific alternative mechanical system. The <|retrieval|> token is inserted to indicate where additional information about gun loading mechanisms might be needed.",
"Ideal Knowledge": "Detailed descriptions of various loading systems used in coastal defense guns, including the Woolwich system and the Armstrong mechanical loader, with their respective advantages and disadvantages.",
"Key Factual Information": "the Armstrong mechanical loader",
"Prefix reformulated": "What was the alternative loading system for the guns besides the manual Woolwich system?",
"Prefix-intent": "The guns were loaded either manually — using the Woolwich ( derrick ) system — or via",
"Verification": "The KFI directly provides the alternative loading system, as prompted by the PI."
},
{
"Explanation": "The PI introduces the crew size and sets up a query about loading time, while the KFI provides the specific duration and compares it to the mechanical system. The <|retrieval|> token is inserted to indicate where additional information about gun loading efficiency might be needed.",
"Ideal Knowledge": "Comparative data on loading times and rates of fire for various coastal defense guns, including manual and mechanical loading systems, with specific information on crew sizes and efficiency.",
"Key Factual Information": "2 minutes , a rate of fire that was only slightly faster using the Armstrong equipment",
"Prefix reformulated": "How long did it take the 14-man crew to load the guns?",
"Prefix-intent": "The guns had a 14 @-@ man crew and loading took about",
"Verification": "The KFI directly provides the loading time and compares it to the Armstrong equipment, as prompted by the PI."
}
]
} |
By 1902 the guns had together fired 219 rounds , though not all at full charge with the records showing only 125 effective full charges . The platforms and traverses were scrapped in 1937 but the gun barrels were uneconomic to cut up and were left in place . One gun platform and its equipment was rebuilt in 1997 using money raised by Fort Glanville Historical Association volunteers working at the Australian Grand Prix .
| [
{
"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 1902 the guns had together fired 219 rounds , though not all at full charge with the records showing only 125 effective full charges . The platforms and traverses were scrapped in 1937 but the gun barrels were uneconomic to cut up and were left in place . One gun platform and its equipment was rebuilt in 1997 using money raised by Fort Glanville Historical Association volunteers working at the Australian Grand Prix . \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|>By 1902 the guns had together fired<|end-PI|> <|retrieval|> <|start-KFI|>219 rounds , though not all at full charge with the records showing only 125 effective full charges<|end-KFI|> . <|start-PI|>The platforms and traverses were scrapped in<|end-PI|> <|retrieval|> <|start-KFI|>1937 but the gun barrels were uneconomic to cut up and were left in place<|end-KFI|> . <|start-PI|>One gun platform and its equipment was rebuilt in<|end-PI|> <|retrieval|> <|start-KFI|>1997 using money raised by Fort Glanville Historical Association volunteers working at the Australian Grand Prix<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI sets up a query about the number of rounds fired by the guns, and the KFI provides the specific count and additional context. The <|retrieval|> token is inserted to indicate where additional information about the firing history might be needed.",
"Ideal Knowledge": "Detailed firing records of the guns at Fort Glanville, including the total number of rounds fired and the breakdown of full charges versus partial charges up to 1902.",
"Key Factual Information": "219 rounds , though not all at full charge with the records showing only 125 effective full charges",
"Prefix reformulated": "How many rounds had the guns at Fort Glanville fired by 1902?",
"Prefix-intent": "By 1902 the guns had together fired",
"Verification": "The KFI directly provides the number of rounds fired and additional details about the charges, as prompted by the PI."
},
{
"Explanation": "The PI introduces a query about when certain parts of the fort were scrapped, and the KFI provides the specific year and additional context about the gun barrels. The <|retrieval|> token is inserted to indicate where additional historical information might be needed.",
"Ideal Knowledge": "Historical records of Fort Glanville's decommissioning process, including dates of when different parts of the fort were scrapped or preserved, with specific focus on the platforms, traverses, and gun barrels.",
"Key Factual Information": "1937 but the gun barrels were uneconomic to cut up and were left in place",
"Prefix reformulated": "When were the platforms and traverses at Fort Glanville scrapped?",
"Prefix-intent": "The platforms and traverses were scrapped in",
"Verification": "The KFI directly provides the year when the platforms and traverses were scrapped and additional information about the gun barrels, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the rebuilding of a gun platform, and the KFI provides the specific year and details about how it was funded. The <|retrieval|> token is inserted to indicate where additional information about the restoration project might be needed.",
"Ideal Knowledge": "Information about the restoration efforts at Fort Glanville, including specific projects, dates, funding sources, and the involvement of historical associations and volunteers.",
"Key Factual Information": "1997 using money raised by Fort Glanville Historical Association volunteers working at the Australian Grand Prix",
"Prefix reformulated": "When was one gun platform and its equipment at Fort Glanville rebuilt?",
"Prefix-intent": "One gun platform and its equipment was rebuilt in",
"Verification": "The KFI directly provides the year of rebuilding and details about the funding source, as prompted by the PI."
}
]
} |
In the flank barbettes are mounted two mark III RML 64 pounder 64 cwt guns , numbers 462 and 463 . They are rifled muzzle @-@ loading heavy guns weighing 64 long hundredweight ( 3 @,@ 300 kg ) with a 6 @.@ 3 @-@ inch ( 160 mm ) calibre steel barrel . They were made to the model 's 1867 pattern by the Royal Arsenal , Woolwich , England in 1872 . Their intended use was to protect the flanks and approaches to the fort . These smaller guns use an 8 lb ( 4 kg ) charge of gunpowder to send a 64 lb ( 29 kg ) projectile up to 5 @,@ 000 yards ( 4 @,@ 600 m ) . There are iron rings fixed in the rampart wall that indicate traversing gun carriages were intended to be mounted , though this never eventuated . The guns were supplied with siege over @-@ bank carriages . The lower carriage sections are the type made for the guns when used as field pieces and the upper sections are brackets that raise the guns to the over @-@ bank firing position . In this particular configuration the two guns are thought to be the last in the world .
| [
{
"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 the flank barbettes are mounted two mark III RML 64 pounder 64 cwt guns , numbers 462 and 463 . They are rifled muzzle @-@ loading heavy guns weighing 64 long hundredweight ( 3 @,@ 300 kg ) with a 6 @.@ 3 @-@ inch ( 160 mm ) calibre steel barrel . They were made to the model 's 1867 pattern by the Royal Arsenal , Woolwich , England in 1872 . Their intended use was to protect the flanks and approaches to the fort . These smaller guns use an 8 lb ( 4 kg ) charge of gunpowder to send a 64 lb ( 29 kg ) projectile up to 5 @,@ 000 yards ( 4 @,@ 600 m ) . There are iron rings fixed in the rampart wall that indicate traversing gun carriages were intended to be mounted , though this never eventuated . The guns were supplied with siege over @-@ bank carriages . The lower carriage sections are the type made for the guns when used as field pieces and the upper sections are brackets that raise the guns to the over @-@ bank firing position . In this particular configuration the two guns are thought to be the last in the world . \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 the flank barbettes are mounted two mark III RML 64 pounder 64 cwt guns , numbers<|end-PI|> <|retrieval|> <|start-KFI|>462 and 463<|end-KFI|>. <|start-PI|>They are rifled muzzle @-@ loading heavy guns weighing<|end-PI|> <|retrieval|> <|start-KFI|>64 long hundredweight ( 3 @,@ 300 kg ) with a 6 @.@ 3 @-@ inch ( 160 mm ) calibre steel barrel<|end-KFI|>. <|start-PI|>They were made to the model 's 1867 pattern by the Royal Arsenal , Woolwich , England in<|end-PI|> <|retrieval|> <|start-KFI|>1872<|end-KFI|>. Their intended use was to protect the flanks and approaches to the fort. <|start-PI|>These smaller guns use an 8 lb ( 4 kg ) charge of gunpowder to send a 64 lb ( 29 kg ) projectile up to<|end-PI|> <|retrieval|> <|start-KFI|>5 @,@ 000 yards ( 4 @,@ 600 m )<|end-KFI|>. There are iron rings fixed in the rampart wall that indicate traversing gun carriages were intended to be mounted , though this never eventuated . The guns were supplied with siege over @-@ bank carriages . The lower carriage sections are the type made for the guns when used as field pieces and the upper sections are brackets that raise the guns to the over @-@ bank firing position . <|start-PI|>In this particular configuration the two guns are thought to be<|end-PI|> <|retrieval|> <|start-KFI|>the last in the world<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces the guns and sets up a query about their specific numbers, while the KFI provides the exact numbers. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "A detailed inventory of mark III RML 64 pounder 64 cwt guns, including their serial numbers and locations where they are mounted.",
"Key Factual Information": "462 and 463",
"Prefix reformulated": "What are the numbers of the two mark III RML 64 pounder 64 cwt guns mounted in the flank barbettes?",
"Prefix-intent": "In the flank barbettes are mounted two mark III RML 64 pounder 64 cwt guns , numbers",
"Verification": "The KFI directly provides the specific numbers of the guns mentioned in the PI."
},
{
"Explanation": "The PI sets up a query about the weight of the guns, and the KFI provides the specific weight and additional details. The <|retrieval|> token is inserted to indicate where additional technical information might be needed.",
"Ideal Knowledge": "Technical specifications of mark III RML 64 pounder 64 cwt guns, including their weight in various units and detailed barrel measurements.",
"Key Factual Information": "64 long hundredweight ( 3 @,@ 300 kg ) with a 6 @.@ 3 @-@ inch ( 160 mm ) calibre steel barrel",
"Prefix reformulated": "How much do the mark III RML 64 pounder 64 cwt guns weigh?",
"Prefix-intent": "They are rifled muzzle @-@ loading heavy guns weighing",
"Verification": "The KFI directly provides the weight and additional specifications of the guns as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about when and where the guns were made, and the KFI provides the specific year. The <|retrieval|> token is inserted to indicate where additional historical information might be needed.",
"Ideal Knowledge": "Historical records of gun production at the Royal Arsenal in Woolwich, England, specifically detailing the manufacture of mark III RML 64 pounder 64 cwt guns in 1872.",
"Key Factual Information": "1872",
"Prefix reformulated": "When were the mark III RML 64 pounder 64 cwt guns made by the Royal Arsenal, Woolwich, England?",
"Prefix-intent": "They were made to the model 's 1867 pattern by the Royal Arsenal , Woolwich , England in",
"Verification": "The KFI directly provides the year of manufacture as prompted by the PI."
},
{
"Explanation": "The PI describes the gun's ammunition and sets up a query about its range, while the KFI provides the specific maximum range. The <|retrieval|> token is inserted to indicate where additional ballistic information might be needed.",
"Ideal Knowledge": "Ballistic data for mark III RML 64 pounder 64 cwt guns, including charge weights, projectile weights, and maximum ranges under various conditions.",
"Key Factual Information": "5 @,@ 000 yards ( 4 @,@ 600 m )",
"Prefix reformulated": "What is the maximum range of the mark III RML 64 pounder 64 cwt guns?",
"Prefix-intent": "These smaller guns use an 8 lb ( 4 kg ) charge of gunpowder to send a 64 lb ( 29 kg ) projectile up to",
"Verification": "The KFI directly provides the maximum range of the guns as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the significance of the guns' configuration, and the KFI provides information about their unique status. The <|retrieval|> token is inserted to indicate where additional historical context might be needed.",
"Ideal Knowledge": "A comprehensive global inventory of surviving mark III RML 64 pounder 64 cwt guns, highlighting the rarity of this particular configuration and its historical significance.",
"Key Factual Information": "the last in the world",
"Prefix reformulated": "What is the significance of the current configuration of these two mark III RML 64 pounder 64 cwt guns?",
"Prefix-intent": "In this particular configuration the two guns are thought to be",
"Verification": "The KFI directly states the unique status of these guns in their current configuration, as prompted by the PI."
}
]
} |
By the end of their active use in 1902 together they had fired 1540 rounds in practice , though not all at full power as they are recorded as firing under 300 effective full charges . By then the wheels were significantly rotted and the ammunition limbers had become unserviceable . Less wheels , the guns were removed from the fort in 1909 ; Adelaide 's city council then set them in Brougham Gardens , North Adelaide . Both were returned to the fort in 1976 and remounted on restored carriages of the original design . They are the only two guns of their type remaining in Australia . The southern gun ( Number 463 ) fired three blank charges in 1980 to mark the centenary of the first firing at the fort . This same gun is fired regularly by the volunteers of the Fort Glanville Historical Association ( FGHA ) .
| [
{
"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 end of their active use in 1902 together they had fired 1540 rounds in practice , though not all at full power as they are recorded as firing under 300 effective full charges . By then the wheels were significantly rotted and the ammunition limbers had become unserviceable . Less wheels , the guns were removed from the fort in 1909 ; Adelaide 's city council then set them in Brougham Gardens , North Adelaide . Both were returned to the fort in 1976 and remounted on restored carriages of the original design . They are the only two guns of their type remaining in Australia . The southern gun ( Number 463 ) fired three blank charges in 1980 to mark the centenary of the first firing at the fort . This same gun is fired regularly by the volunteers of the Fort Glanville Historical Association ( FGHA ) . \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
} |
Both types of heavy guns used where the only ones of their specific series and type to come to Australia . Though they were never fired in anger , the battery is only 40 feet ( 12 m ) above mean sea level which would have limited the accuracy of the mark III depression rangefinders used .
| [
{
"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 Both types of heavy guns used where the only ones of their specific series and type to come to Australia . Though they were never fired in anger , the battery is only 40 feet ( 12 m ) above mean sea level which would have limited the accuracy of the mark III depression rangefinders used . \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
} |
Prior to 1895 two 6 @-@ inch ( 200 mm ) breech @-@ loading guns had been imported for use in small boats . This use was rejected by military authorities and the guns sat unused . In 1895 the South Australian Defence Committee proposed that the guns be mounted at Fort Glanville , replacing the 64 @-@ pounders whose siege carriages were then unfit for service . This proposal would have greatly extended the useful life of the fort , at little cost . The two military branches , army and navy , could not reach agreement and the proposal was abandoned by 1897 . The gun 's eventual fate is unknown , though a gun found in the Port River later indicates they may have been dumped .
| [
{
"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 Prior to 1895 two 6 @-@ inch ( 200 mm ) breech @-@ loading guns had been imported for use in small boats . This use was rejected by military authorities and the guns sat unused . In 1895 the South Australian Defence Committee proposed that the guns be mounted at Fort Glanville , replacing the 64 @-@ pounders whose siege carriages were then unfit for service . This proposal would have greatly extended the useful life of the fort , at little cost . The two military branches , army and navy , could not reach agreement and the proposal was abandoned by 1897 . The gun 's eventual fate is unknown , though a gun found in the Port River later indicates they may have been dumped . \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
} |
Gun emplacements were made , about a mile apart south of the fort , for six 16 @-@ pounder field guns to provide close defence in case of an attempted beach landing . From 1890 the fort was equipped with moveable armament . It received an 1867 Whitworth 12 @-@ pounder RML field gun and a 5 @-@ barrel 0 @.@ 45 " Martini @-@ henry cartridge firing Nordenfelt gun . The 12 @-@ pounder had fired 222 effective full charges by 1902 ; at which time both weapons were obsolete and their eventual fate is unknown . A 32 @-@ pounder 56 long cwt ( 2 @,@ 800 kg ) smoothbore gun on a stepped wooden carriage was purchased in 1878 . Manufactured by the Royal Arsenal in 1806 , it was the familiar " ships cannon " and came to South Australia with two other identical guns . By 1902 no ammunition was held for it and it was listed as for instruction only . The gun was probably destroyed in the 1930s — certainly by the end of the 1940s — with pieces of such a gun found on site in 1983 .
| [
{
"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 Gun emplacements were made , about a mile apart south of the fort , for six 16 @-@ pounder field guns to provide close defence in case of an attempted beach landing . From 1890 the fort was equipped with moveable armament . It received an 1867 Whitworth 12 @-@ pounder RML field gun and a 5 @-@ barrel 0 @.@ 45 \" Martini @-@ henry cartridge firing Nordenfelt gun . The 12 @-@ pounder had fired 222 effective full charges by 1902 ; at which time both weapons were obsolete and their eventual fate is unknown . A 32 @-@ pounder 56 long cwt ( 2 @,@ 800 kg ) smoothbore gun on a stepped wooden carriage was purchased in 1878 . Manufactured by the Royal Arsenal in 1806 , it was the familiar \" ships cannon \" and came to South Australia with two other identical guns . By 1902 no ammunition was held for it and it was listed as for instruction only . The gun was probably destroyed in the 1930s — certainly by the end of the 1940s — with pieces of such a gun found on site in 1983 . \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 the late 20th century the fort acquired three 16 @-@ pounder RML mark I field guns . They were formerly used by A battery , South Australian Volunteer Artillery from 1880 until 1901 . Gun number 288 is complete and used for blank firings by the Historical Association . Also acquired is a 2 @-@ pounder RML Whitworth mountain gun made in 1867 . It was also used by A Battery , subsequently by Fort Largs as a signal gun . It is one of only two of this type known to exist , the other in the United Kingdom . The visitors centre has two 9 @-@ pounder brass smoothbore field guns made by H & C King in 1819 . They arrived in South Australia in 1857 and were used for practice shoots near the fort c.1860 ; one is known to have been on the manning parade in 1890 though its use is unknown . They later became saluting guns at Fort Largs , moving by 1919 to near the Jervois Wing of the State Library of South Australia . The Art Gallery of South Australia saved them from a 1941 wartime scrap drive and mounted them on reproduction naval carriages in front of Government House in 1962 . The gallery took them back in late 1977 and transferred them to the History Trust of South Australia in 1988 . The History Trust has loaned them to Fort Glanville for display . Outside the visitor 's centre is a 6 in ( 200 mm ) breech @-@ loading Armstrong 80 long cwt ( 4 @,@ 100 kg ) gun ( No.4242 ) that was made in 1884 and used in Victoria . The Commonwealth Scientific and Industrial Research Organisation ( CSIRO ) brought it to the state in 1966 for research , subsequently moving it to Perry Engineering at Mile End . In 1984 the CSIRO donated the gun to the 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 In the late 20th century the fort acquired three 16 @-@ pounder RML mark I field guns . They were formerly used by A battery , South Australian Volunteer Artillery from 1880 until 1901 . Gun number 288 is complete and used for blank firings by the Historical Association . Also acquired is a 2 @-@ pounder RML Whitworth mountain gun made in 1867 . It was also used by A Battery , subsequently by Fort Largs as a signal gun . It is one of only two of this type known to exist , the other in the United Kingdom . The visitors centre has two 9 @-@ pounder brass smoothbore field guns made by H & C King in 1819 . They arrived in South Australia in 1857 and were used for practice shoots near the fort c.1860 ; one is known to have been on the manning parade in 1890 though its use is unknown . They later became saluting guns at Fort Largs , moving by 1919 to near the Jervois Wing of the State Library of South Australia . The Art Gallery of South Australia saved them from a 1941 wartime scrap drive and mounted them on reproduction naval carriages in front of Government House in 1962 . The gallery took them back in late 1977 and transferred them to the History Trust of South Australia in 1988 . The History Trust has loaned them to Fort Glanville for display . Outside the visitor 's centre is a 6 in ( 200 mm ) breech @-@ loading Armstrong 80 long cwt ( 4 @,@ 100 kg ) gun ( No.4242 ) that was made in 1884 and used in Victoria . The Commonwealth Scientific and Industrial Research Organisation ( CSIRO ) brought it to the state in 1966 for research , subsequently moving it to Perry Engineering at Mile End . In 1984 the CSIRO donated the gun to the 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
} |
Work began in 1882 on Fort Glanville 's northern sister Fort Largs — then known as the Port Adelaide battery — to the same specification as Fort Glanville ; its barracks and rear defence wall were finished in 1885 . Though guns had already been purchased specially , plans for the third fort at Glenelg were not proceeded with . South Australia was experiencing a depression in 1886 and that coupled with a report by General James Bevan Edwards scuttled plans for the third coastal fort . As early as 1888 the emphasis for defence of the Adelaide coast had already shifted to Fort Largs ; a fact cited as part of the reason for abandonment of the Glenelg fortifications .
| [
{
"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 Work began in 1882 on Fort Glanville 's northern sister Fort Largs — then known as the Port Adelaide battery — to the same specification as Fort Glanville ; its barracks and rear defence wall were finished in 1885 . Though guns had already been purchased specially , plans for the third fort at Glenelg were not proceeded with . South Australia was experiencing a depression in 1886 and that coupled with a report by General James Bevan Edwards scuttled plans for the third coastal fort . As early as 1888 the emphasis for defence of the Adelaide coast had already shifted to Fort Largs ; a fact cited as part of the reason for abandonment of the Glenelg fortifications . \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
} |
From this point Fort Glanville 's significance declined rapidly . Fort Largs was equipped in 1889 with two 6 inch breech @-@ loading disappearing guns which outranged Glanville 's armament . The decline was also linked to changes in Port Adelaide 's maritime facilities and the consequent northward movement of anchored and berthed vessels . During the 1880s the Port River was deepened enabling large ships to sail up and berth , rather than the former practice of anchoring off @-@ shore .
| [
{
"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 From this point Fort Glanville 's significance declined rapidly . Fort Largs was equipped in 1889 with two 6 inch breech @-@ loading disappearing guns which outranged Glanville 's armament . The decline was also linked to changes in Port Adelaide 's maritime facilities and the consequent northward movement of anchored and berthed vessels . During the 1880s the Port River was deepened enabling large ships to sail up and berth , rather than the former practice of anchoring off @-@ shore . \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 fort remained as headquarters of South Australia 's permanent military force until the 1890s and as late as 1895 there were still plans to upgrade the 64 @-@ pounder armament , though without result . By 1901 the fort was manned on a caretaker basis only and no permanent forces were stationed . The Federal Government assumed responsibility for South Australian defence in 1903 and took over the fort . Though Glanville by then had no significant defence role , the state received £ 14 @,@ 739 in compensation . From that point its significance was not defence related but as the first and best preserved 19th @-@ century fortification in South Australia .
| [
{
"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 fort remained as headquarters of South Australia 's permanent military force until the 1890s and as late as 1895 there were still plans to upgrade the 64 @-@ pounder armament , though without result . By 1901 the fort was manned on a caretaker basis only and no permanent forces were stationed . The Federal Government assumed responsibility for South Australian defence in 1903 and took over the fort . Though Glanville by then had no significant defence role , the state received £ 14 @,@ 739 in compensation . From that point its significance was not defence related but as the first and best preserved 19th @-@ century fortification in South Australia . \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
} |
At Federation in 1901 South Australia 's defences became a federal responsibility . Though both infantry and artillery units were housed at the site on occasion , by 1903 there was no longer a permanent military presence at the fort . For most of the 20th century the site was neglected and largely vacated . It did attract some usage though not always of a military nature . During World War I it was partly revived for a former use , with ammunition stored on site . During the same period however the military used it as a detention barracks . Some or all of the fort was leased for private accommodation during the great depression . During the 1930s the magazine was again used , this time to store small arms ammunition . From June 1931 until the beginning of World War II the site housed a Sea Scout detachment , and was used as a district camp @-@ site for the Boy Scouts . The Department of Defence decided in 1937 that much of the equipment and fittings at the fort were surplus to requirements . Consequent to this the mountings and carriages for the 10 inch guns were removed and sold as scrap ; though effort was made to scrap the barrels it proved uneconomic and they were left in place . In an unusual turn of events , the fort briefly housed refugees . Twenty @-@ nine islanders from the Maldives were rescued from their sinking dhow by a ship bound for Wallaroo . They stayed at the fort for a week in 1938 , before repatriation could be arranged . During World War II the fort again attracted military related use . During 1944 the Proof and Experimental Establishment at Port Wakefield made use of the site to proof ordnance QF 6 pounder anti @-@ tank guns manufactured at General Motors Holden in Woodville . The fort was also used as a residence again , with at least two families reported as living in the under @-@ ground sections during the war .
| [
{
"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 Federation in 1901 South Australia 's defences became a federal responsibility . Though both infantry and artillery units were housed at the site on occasion , by 1903 there was no longer a permanent military presence at the fort . For most of the 20th century the site was neglected and largely vacated . It did attract some usage though not always of a military nature . During World War I it was partly revived for a former use , with ammunition stored on site . During the same period however the military used it as a detention barracks . Some or all of the fort was leased for private accommodation during the great depression . During the 1930s the magazine was again used , this time to store small arms ammunition . From June 1931 until the beginning of World War II the site housed a Sea Scout detachment , and was used as a district camp @-@ site for the Boy Scouts . The Department of Defence decided in 1937 that much of the equipment and fittings at the fort were surplus to requirements . Consequent to this the mountings and carriages for the 10 inch guns were removed and sold as scrap ; though effort was made to scrap the barrels it proved uneconomic and they were left in place . In an unusual turn of events , the fort briefly housed refugees . Twenty @-@ nine islanders from the Maldives were rescued from their sinking dhow by a ship bound for Wallaroo . They stayed at the fort for a week in 1938 , before repatriation could be arranged . During World War II the fort again attracted military related use . During 1944 the Proof and Experimental Establishment at Port Wakefield made use of the site to proof ordnance QF 6 pounder anti @-@ tank guns manufactured at General Motors Holden in Woodville . The fort was also used as a residence again , with at least two families reported as living in the under @-@ ground sections during the war . \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 the war , the State Government negotiated with its federal counterpart to gain control of Fort Glanville . This eventuated in the 1951 sale of the 13 @-@ acre ( 5 ha ) site , which was subsequently administered by the State Tourist Department as a caravan and camping park . The park occupied the muster ground outside the fort and the fort 's barracks building was used as a manager 's residence . The caravan park has had various managers since establishment : Until 1981 it was managed by the National Parks and Wildlife Service ( NPWS ) , by Woodville Council until 1986 and subsequently by a private operator under a long @-@ term lease .
| [
{
"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 the war , the State Government negotiated with its federal counterpart to gain control of Fort Glanville . This eventuated in the 1951 sale of the 13 @-@ acre ( 5 ha ) site , which was subsequently administered by the State Tourist Department as a caravan and camping park . The park occupied the muster ground outside the fort and the fort 's barracks building was used as a manager 's residence . The caravan park has had various managers since establishment : Until 1981 it was managed by the National Parks and Wildlife Service ( NPWS ) , by Woodville Council until 1986 and subsequently by a private operator under a long @-@ term lease . \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 significant change in the fort 's conservation and preservation outlook was its declaration as a Historic Relic under the Aboriginal and Historic Relics Preservation Act in 1972 . Up to then there had been growing awareness of the significance of the site in terms of the state 's colonial heritage . Control of the park was moved to the National Parks and Wildlife Service ( NPWS ) , and the fort and surrounds became designated as Fort Glanville Conservation Park . The National Estate Grants Program provided funds in 1975 for conservation work ; NPWS began this work in the same year and the caravan park boundary was moved southwards . Though the site had long been neglected , there was minimal permanent damage to its structure . During the 1970s reconstruction , much of the stockade was replaced . The replaced timbers are visibly different as they have shrunk significantly with large gaps that are not evident in the original timber . Adelaide City Council agreed to return the 64 @-@ pounder guns and new wheels were made ; replacing those rotted over half a century before . In a ceremony on 2 October 1980 , the 100th anniversary of the guns first firing , one of the 64 @-@ pounders was fired again ; later a commemoration plaque was added to the fort 's flagpole 's base . Shortly after this firing the Fort Glanville Historical Association was formed , and was incorporated in 1981 . Fort Glanville was opened for public visitation in 1981 .
| [
{
"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 significant change in the fort 's conservation and preservation outlook was its declaration as a Historic Relic under the Aboriginal and Historic Relics Preservation Act in 1972 . Up to then there had been growing awareness of the significance of the site in terms of the state 's colonial heritage . Control of the park was moved to the National Parks and Wildlife Service ( NPWS ) , and the fort and surrounds became designated as Fort Glanville Conservation Park . The National Estate Grants Program provided funds in 1975 for conservation work ; NPWS began this work in the same year and the caravan park boundary was moved southwards . Though the site had long been neglected , there was minimal permanent damage to its structure . During the 1970s reconstruction , much of the stockade was replaced . The replaced timbers are visibly different as they have shrunk significantly with large gaps that are not evident in the original timber . Adelaide City Council agreed to return the 64 @-@ pounder guns and new wheels were made ; replacing those rotted over half a century before . In a ceremony on 2 October 1980 , the 100th anniversary of the guns first firing , one of the 64 @-@ pounders was fired again ; later a commemoration plaque was added to the fort 's flagpole 's base . Shortly after this firing the Fort Glanville Historical Association was formed , and was incorporated in 1981 . Fort Glanville was opened for public visitation in 1981 . \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 part of South Australia 's sesquicentenary , the South Australia Jubilee 150 board granted $ 250 @,@ 000 for work on the fort , allowing construction of a visitor centre . Queen Elizabeth II visited and inspected the site , and newly constructed visitor centre , for almost an hour on 13 March 1986 . For the visit the historical association demonstrated firing of both the 64 and 2 pounder cannons . The visitor centre was officially opened one month after this visit . Governor Dame Roma Mitchell named the old road " Queen Elizabeth II Walk " in March 1991 ceremony , commemorating the 5th anniversary of the queen 's visit . Considerable subsequent restoration work has been completed on the fort , including the barrack 's interior . The portion of the muster ground that is not within the caravan park was returned to its original level in 1993 .
| [
{
"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 part of South Australia 's sesquicentenary , the South Australia Jubilee 150 board granted $ 250 @,@ 000 for work on the fort , allowing construction of a visitor centre . Queen Elizabeth II visited and inspected the site , and newly constructed visitor centre , for almost an hour on 13 March 1986 . For the visit the historical association demonstrated firing of both the 64 and 2 pounder cannons . The visitor centre was officially opened one month after this visit . Governor Dame Roma Mitchell named the old road \" Queen Elizabeth II Walk \" in March 1991 ceremony , commemorating the 5th anniversary of the queen 's visit . Considerable subsequent restoration work has been completed on the fort , including the barrack 's interior . The portion of the muster ground that is not within the caravan park was returned to its original level in 1993 . \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
} |
Fort Glanville Conservation Park is one of South Australia 's most important heritage sites . It is the premier site in the state , and possibly Australia , for showcasing colonial era defences and fortifications . The fort is listed on both the South Australian Heritage Register and the National Trust 's classified list . It is considered significant , in national historical military terms , for its association with Jervois and Scratchley ; both leading British defence experts who influenced Australian defence thinking in the late 19th century . The fort was very well constructed with high quality materials ; facts that have been noted as contributing to its preservation . The fort is largely intact and in original condition and is , along with Bluff Battery in Hobart , the best preserved Jervois @-@ Scratchley designed fort in Australia . It is the only Australian colonial fort to still have all of its original armament , and the only to have a regular living history program . Its companion Fort Largs does survive but in greatly modified condition , converted for use as the South Australia Police academy . The fort retains its original armament ; armament that is both rare in Australia and worldwide .
| [
{
"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 Fort Glanville Conservation Park is one of South Australia 's most important heritage sites . It is the premier site in the state , and possibly Australia , for showcasing colonial era defences and fortifications . The fort is listed on both the South Australian Heritage Register and the National Trust 's classified list . It is considered significant , in national historical military terms , for its association with Jervois and Scratchley ; both leading British defence experts who influenced Australian defence thinking in the late 19th century . The fort was very well constructed with high quality materials ; facts that have been noted as contributing to its preservation . The fort is largely intact and in original condition and is , along with Bluff Battery in Hobart , the best preserved Jervois @-@ Scratchley designed fort in Australia . It is the only Australian colonial fort to still have all of its original armament , and the only to have a regular living history program . Its companion Fort Largs does survive but in greatly modified condition , converted for use as the South Australia Police academy . The fort retains its original armament ; armament that is both rare in Australia and worldwide . \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 conservation park is in the suburb of Semaphore Park at the southern end of the Lefevre Peninsula . It covers approximately 5 hectares ( 12 acres ) and is bounded by Bower and Military roads , the Point Malcolm Reserve and Semaphore beach . The land is entirely crown land and is administered by the DEWNR . It is divided into roughly two halves with the northern containing the fort and the southern half containing a caravan park . While it is managed by the DEWNR , the Fort Glanville Consultative Committee and Fort Glanville Historical Association are heavily involved . The committee is appointed by the Minister for Environment and Planning and acts to advise the Minister and liaise with the community . As of 1988 , the fort hosted approximately 7 @,@ 000 visitors annually , and is used as an event venue including the annual City of Charles Sturt citizenship ceremony .
| [
{
"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 conservation park is in the suburb of Semaphore Park at the southern end of the Lefevre Peninsula . It covers approximately 5 hectares ( 12 acres ) and is bounded by Bower and Military roads , the Point Malcolm Reserve and Semaphore beach . The land is entirely crown land and is administered by the DEWNR . It is divided into roughly two halves with the northern containing the fort and the southern half containing a caravan park . While it is managed by the DEWNR , the Fort Glanville Consultative Committee and Fort Glanville Historical Association are heavily involved . The committee is appointed by the Minister for Environment and Planning and acts to advise the Minister and liaise with the community . As of 1988 , the fort hosted approximately 7 @,@ 000 visitors annually , and is used as an event venue including the annual City of Charles Sturt citizenship ceremony . \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 conservation park is in the suburb of Semaphore Park at<|end-PI|> <|retrieval|> <|start-KFI|>the southern end of the Lefevre Peninsula<|end-KFI|>. <|start-PI|>It covers approximately<|end-PI|> <|retrieval|> <|start-KFI|>5 hectares ( 12 acres )<|end-KFI|> and is bounded by Bower and Military roads , the Point Malcolm Reserve and Semaphore beach . <|start-PI|>The land is entirely crown land and is administered by<|end-PI|> <|retrieval|> <|start-KFI|>the DEWNR<|end-KFI|> . It is divided into roughly two halves with the northern containing the fort and the southern half containing a caravan park . While it is managed by the DEWNR , the Fort Glanville Consultative Committee and Fort Glanville Historical Association are heavily involved . The committee is appointed by the Minister for Environment and Planning and acts to advise the Minister and liaise with the community . <|start-PI|>As of 1988 , the fort hosted approximately<|end-PI|> <|retrieval|> <|start-KFI|>7 @,@ 000 visitors annually<|end-KFI|> , and is used as an event venue including the annual City of Charles Sturt citizenship ceremony .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the conservation park and sets up a query about its location, while the KFI provides the specific geographical information. The <|retrieval|> token is inserted to indicate where additional location details might be needed.",
"Ideal Knowledge": "Detailed geographical information about Semaphore Park and the Lefevre Peninsula, including their relative positions and notable landmarks.",
"Key Factual Information": "the southern end of the Lefevre Peninsula",
"Prefix reformulated": "Where is the conservation park located?",
"Prefix-intent": "The conservation park is in the suburb of Semaphore Park at",
"Verification": "The KFI directly provides the location information requested in the PI."
},
{
"Explanation": "The PI sets up a query about the park's size, and the KFI provides the specific measurements. The <|retrieval|> token is inserted to indicate where additional information about park sizes might be needed.",
"Ideal Knowledge": "A list of conservation parks in the area with their sizes, specifically mentioning this park's area in both hectares and acres.",
"Key Factual Information": "5 hectares ( 12 acres )",
"Prefix reformulated": "What is the approximate size of the conservation park?",
"Prefix-intent": "It covers approximately",
"Verification": "The KFI directly provides the size information requested in the PI."
},
{
"Explanation": "The PI sets up a query about the park's administration, and the KFI provides the specific organization. The <|retrieval|> token is inserted to indicate where additional information about the administrative body might be needed.",
"Ideal Knowledge": "Information about the DEWNR (Department of Environment, Water and Natural Resources), its role in managing crown lands, and other conservation areas under its administration.",
"Key Factual Information": "the DEWNR",
"Prefix reformulated": "Who administers the conservation park?",
"Prefix-intent": "The land is entirely crown land and is administered by",
"Verification": "The KFI directly provides the administrative body as requested in the PI."
},
{
"Explanation": "The PI sets up a query about the fort's visitor numbers, and the KFI provides the specific annual count. The <|retrieval|> token is inserted to indicate where additional information about visitor statistics might be needed.",
"Ideal Knowledge": "Historical visitor statistics for Fort Glanville, including trends over time and comparisons with other similar attractions in the area.",
"Key Factual Information": "7 @,@ 000 visitors annually",
"Prefix reformulated": "How many visitors did the fort host annually as of 1988?",
"Prefix-intent": "As of 1988 , the fort hosted approximately",
"Verification": "The KFI directly provides the visitor count information requested in the PI."
}
]
} |
The historical association is an incorporated body of volunteers who use the fort under licence from the Minister . The association operate the fort and visitor centre , holding monthly public open days at which the history of the fort is recreated ; this includes drill demonstrations and firing of the fort 's weapons . The association maintain static displays of the fort 's active period and sell souvenirs . The association 's goal is to have the site open as a fully operational fort for interpretation , tourism and education . The volunteers are involved with recreation , living history and ceremonial work outside the fort . They act as guards of honour at some Government House functions and attended the restoration opening of the Albany , Western Australia fort , and the annual re @-@ enactment of the Battle of Waterloo in Keyneton , Victoria . The fort has living history displays , including the barracks laid out and furnished in period detail . The visitor centre has displays showing the development of South Australia 's colonial defence from 1836 and artifacts found on site during restorations .
| [
{
"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 historical association is an incorporated body of volunteers who use the fort under licence from the Minister . The association operate the fort and visitor centre , holding monthly public open days at which the history of the fort is recreated ; this includes drill demonstrations and firing of the fort 's weapons . The association maintain static displays of the fort 's active period and sell souvenirs . The association 's goal is to have the site open as a fully operational fort for interpretation , tourism and education . The volunteers are involved with recreation , living history and ceremonial work outside the fort . They act as guards of honour at some Government House functions and attended the restoration opening of the Albany , Western Australia fort , and the annual re @-@ enactment of the Battle of Waterloo in Keyneton , Victoria . The fort has living history displays , including the barracks laid out and furnished in period detail . The visitor centre has displays showing the development of South Australia 's colonial defence from 1836 and artifacts found on site during restorations . \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
} |
Connecting the fort to Semaphore jetty is the Semaphore and Fort Glanville Tourist Railway , a 457 mm ( 18 @.@ 0 in ) gauge steam train operated by volunteers from the National Railway Museum . The railway opened in December 1992 and , as of 2002 , carried over 16 @,@ 000 passengers annually .
| [
{
"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 Connecting the fort to Semaphore jetty is the Semaphore and Fort Glanville Tourist Railway , a 457 mm ( 18 @.@ 0 in ) gauge steam train operated by volunteers from the National Railway Museum . The railway opened in December 1992 and , as of 2002 , carried over 16 @,@ 000 passengers annually . \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": "Connecting the fort to Semaphore jetty is <|start-PI|> the Semaphore and Fort Glanville Tourist Railway , a <|end-PI|> <|retrieval|> <|start-KFI|> 457 mm ( 18 @.@ 0 in ) gauge steam train <|end-KFI|> operated by volunteers from the National Railway Museum . <|start-PI|> The railway opened in <|end-PI|> <|retrieval|> <|start-KFI|> December 1992 <|end-KFI|> and , <|start-PI|> as of 2002 , carried over <|end-PI|> <|retrieval|> <|start-KFI|> 16 @,@ 000 passengers annually <|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the railway and sets up a query about its specifications, while the KFI provides the specific gauge measurement. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "A detailed list of tourist railways in Australia, specifically mentioning the Semaphore and Fort Glanville Tourist Railway, its gauge, and type of train used.",
"Key Factual Information": "457 mm ( 18 @.@ 0 in ) gauge steam train",
"Prefix reformulated": "What is the gauge of the Semaphore and Fort Glanville Tourist Railway?",
"Prefix-intent": "Connecting the fort to Semaphore jetty is the Semaphore and Fort Glanville Tourist Railway , a",
"Verification": "The KFI directly provides the gauge information requested in the PI."
},
{
"Explanation": "The PI sets up a query about the railway's opening date, and the KFI provides the specific month and year. The <|retrieval|> token is inserted to indicate where additional information about the railway's history might be needed.",
"Ideal Knowledge": "A timeline of tourist railway openings in Australia, with specific focus on the Semaphore and Fort Glanville Tourist Railway, including its opening date and any significant events or milestones since then.",
"Key Factual Information": "December 1992",
"Prefix reformulated": "When did the Semaphore and Fort Glanville Tourist Railway open?",
"Prefix-intent": "The railway opened in",
"Verification": "The KFI directly provides the opening date information requested in the PI."
},
{
"Explanation": "The PI sets up a query about the railway's passenger numbers as of a specific year, and the KFI provides the exact figure. The <|retrieval|> token is inserted to indicate where additional information about passenger statistics might be needed.",
"Ideal Knowledge": "Statistical data on passenger numbers for tourist railways in Australia, with a focus on the Semaphore and Fort Glanville Tourist Railway, including yearly passenger counts and any trends or changes over time.",
"Key Factual Information": "16 @,@ 000 passengers annually",
"Prefix reformulated": "How many passengers did the Semaphore and Fort Glanville Tourist Railway carry annually as of 2002?",
"Prefix-intent": "as of 2002 , carried over",
"Verification": "The KFI directly provides the annual passenger count information requested in the PI."
}
]
} |
The royal prerogative is a body of customary authority , privilege , and immunity , recognised in the United Kingdom as the sole prerogative of the Sovereign and the source of many of the executive powers of the British government .
| [
{
"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 royal prerogative is a body of customary authority , privilege , and immunity , recognised in the United Kingdom as the sole prerogative of the Sovereign and the source of many of the executive powers of the British government . \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 royal prerogative is a body of customary authority , privilege , and immunity , recognised in the United Kingdom as<|end-PI|> <|retrieval|> <|start-KFI|>the sole prerogative of the Sovereign and the source of many of the executive powers of the British government<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces the concept of the royal prerogative and sets up a query about its recognition in the UK. The KFI provides specific information about its status and role. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement about the royal prerogative's recognition and function.",
"Ideal Knowledge": "A detailed explanation of the royal prerogative in the UK, including its definition, historical context, and its role in the British government system. This should include information on how it relates to the Sovereign's powers and its influence on executive powers.",
"Key Factual Information": "the sole prerogative of the Sovereign and the source of many of the executive powers of the British government",
"Prefix reformulated": "What is the royal prerogative recognized as in the United Kingdom?",
"Prefix-intent": "The royal prerogative is a body of customary authority , privilege , and immunity , recognised in the United Kingdom as",
"Verification": "The KFI directly provides information about what the royal prerogative is recognized as in the UK, as prompted by the PI."
}
]
} |
Prerogative powers were formerly exercised by the monarch acting on his or her own initiative . Since the 19th century , by convention , the advice of the prime minister or the cabinet — who are then accountable to Parliament for the decision — has been required in order for the prerogative to be exercised . The monarch remains constitutionally empowered to exercise the royal prerogative against the advice of the prime minister or the cabinet , but in practice would only do so in emergencies or where existing precedent does not adequately apply to the circumstances in question .
| [
{
"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 Prerogative powers were formerly exercised by the monarch acting on his or her own initiative . Since the 19th century , by convention , the advice of the prime minister or the cabinet — who are then accountable to Parliament for the decision — has been required in order for the prerogative to be exercised . The monarch remains constitutionally empowered to exercise the royal prerogative against the advice of the prime minister or the cabinet , but in practice would only do so in emergencies or where existing precedent does not adequately apply to the circumstances in question . \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|>Prerogative powers were formerly exercised by<|end-PI|> <|retrieval|> <|start-KFI|>the monarch acting on his or her own initiative<|end-KFI|> . <|start-PI|>Since the 19th century , by convention , the advice of<|end-PI|> <|retrieval|> <|start-KFI|>the prime minister or the cabinet — who are then accountable to Parliament for the decision — has been required in order for the prerogative to be exercised<|end-KFI|> . <|start-PI|>The monarch remains constitutionally empowered to exercise the royal prerogative against the advice of the prime minister or the cabinet , but in practice would only do so<|end-PI|> <|retrieval|> <|start-KFI|>in emergencies or where existing precedent does not adequately apply to the circumstances in question<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the concept of prerogative powers and sets up a query about who exercised them, while the KFI provides the specific answer. The <|retrieval|> token is inserted to indicate where additional historical context might be needed.",
"Ideal Knowledge": "Historical information about the role of monarchs in exercising prerogative powers, including examples of monarchs who acted on their own initiative.",
"Key Factual Information": "the monarch acting on his or her own initiative",
"Prefix reformulated": "Who formerly exercised prerogative powers?",
"Prefix-intent": "Prerogative powers were formerly exercised by",
"Verification": "The KFI directly answers who exercised prerogative powers, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the change in convention since the 19th century, and the KFI provides the specific information about who now advises on prerogative powers. The <|retrieval|> token is inserted to indicate where additional information about this constitutional change might be needed.",
"Ideal Knowledge": "Detailed explanation of the constitutional changes in the 19th century regarding the exercise of prerogative powers, including the roles of the prime minister, cabinet, and Parliament.",
"Key Factual Information": "the prime minister or the cabinet — who are then accountable to Parliament for the decision — has been required in order for the prerogative to be exercised",
"Prefix reformulated": "Whose advice has been required for exercising prerogative powers since the 19th century?",
"Prefix-intent": "Since the 19th century , by convention , the advice of",
"Verification": "The KFI directly provides information about whose advice is required and their accountability, as prompted by the PI."
},
{
"Explanation": "The PI introduces the monarch's constitutional power and sets up a query about when it would be used, while the KFI provides the specific circumstances. The <|retrieval|> token is inserted to indicate where additional information about these exceptional circumstances might be needed.",
"Ideal Knowledge": "Examples of historical emergencies or unprecedented situations where a monarch has exercised or considered exercising the royal prerogative against advice, along with the constitutional implications of such actions.",
"Key Factual Information": "in emergencies or where existing precedent does not adequately apply to the circumstances in question",
"Prefix reformulated": "When would the monarch exercise the royal prerogative against advice?",
"Prefix-intent": "The monarch remains constitutionally empowered to exercise the royal prerogative against the advice of the prime minister or the cabinet , but in practice would only do so",
"Verification": "The KFI directly provides the circumstances under which the monarch would exercise the royal prerogative against advice, as prompted by the PI."
}
]
} |
Today the royal prerogative is available in the conduct of the government of the United Kingdom , including foreign affairs , defence , and national security . The monarchy has a significant constitutional presence in these and other matters , but limited power , because the exercise of the prerogative is in the hands of the prime minister and other ministers or other government officials .
| [
{
"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 Today the royal prerogative is available in the conduct of the government of the United Kingdom , including foreign affairs , defence , and national security . The monarchy has a significant constitutional presence in these and other matters , but limited power , because the exercise of the prerogative is in the hands of the prime minister and other ministers or other government officials . \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|> Today the royal prerogative is available in the conduct of the government of the United Kingdom , including <|end-PI|> <|retrieval|> <|start-KFI|> foreign affairs , defence , and national security <|end-KFI|> . <|start-PI|> The monarchy has a significant constitutional presence in these and other matters , but <|end-PI|> <|retrieval|> <|start-KFI|> limited power , because the exercise of the prerogative is in the hands of the prime minister and other ministers or other government officials <|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the concept of royal prerogative in UK governance and sets up a query about its scope, while the KFI provides specific areas of application. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "A comprehensive list of areas where the royal prerogative applies in the UK government, with specific focus on foreign affairs, defence, and national security, including historical context and current applications.",
"Key Factual Information": "foreign affairs , defence , and national security",
"Prefix reformulated": "What areas does the royal prerogative cover in the conduct of the UK government?",
"Prefix-intent": "Today the royal prerogative is available in the conduct of the government of the United Kingdom , including",
"Verification": "The KFI directly lists the areas covered by the royal prerogative, as prompted by the PI."
},
{
"Explanation": "The PI acknowledges the monarchy's constitutional presence but sets up a query about its actual power, while the KFI provides the explanation of limited power and its delegation. The <|retrieval|> token is inserted to indicate where additional information about the constitutional balance of power might be needed.",
"Ideal Knowledge": "Detailed explanation of the constitutional role of the monarchy in the UK, including the concept of limited power and the delegation of authority to the prime minister and other government officials. This should include historical context and specific examples of how this system operates in practice.",
"Key Factual Information": "limited power , because the exercise of the prerogative is in the hands of the prime minister and other ministers or other government officials",
"Prefix reformulated": "What is the extent of the monarchy's power in constitutional matters?",
"Prefix-intent": "The monarchy has a significant constitutional presence in these and other matters , but",
"Verification": "The KFI directly explains the limitation of the monarchy's power and why, as prompted by the PI."
}
]
} |
The royal prerogative has been called " a notoriously difficult concept to define adequately " , but whether a particular type of prerogative power exists is a matter of common law to be decided by the courts as the final arbiter . A prominent constitutional theorist , A. V. Dicey , proposed in the nineteenth century 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 The royal prerogative has been called \" a notoriously difficult concept to define adequately \" , but whether a particular type of prerogative power exists is a matter of common law to be decided by the courts as the final arbiter . A prominent constitutional theorist , A. V. Dicey , proposed in the nineteenth century 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": null,
"none": "",
"results": null
} |
The prerogative appears to be historically and as a matter of fact nothing else than the residue of discretionary or arbitrary authority which at any given time is legally left in the hands of the crown . The prerogative is the name of the remaining portion of the Crown 's original authority ... Every act which the executive government can lawfully do without the authority of an Act of Parliament is done in virtue of the prerogative .
| [
{
"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 prerogative appears to be historically and as a matter of fact nothing else than the residue of discretionary or arbitrary authority which at any given time is legally left in the hands of the crown . The prerogative is the name of the remaining portion of the Crown 's original authority ... Every act which the executive government can lawfully do without the authority of an Act of Parliament is done in virtue of the prerogative . \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 prerogative appears to be historically and as a matter of fact<|end-PI|> <|retrieval|> <|start-KFI|>nothing else than the residue of discretionary or arbitrary authority which at any given time is legally left in the hands of the crown<|end-KFI|>. <|start-PI|>The prerogative is the name of<|end-PI|> <|retrieval|> <|start-KFI|>the remaining portion of the Crown's original authority<|end-KFI|> ... <|start-PI|>Every act which the executive government can lawfully do without the authority of an Act of Parliament is done in virtue of<|end-PI|> <|retrieval|> <|start-KFI|>the prerogative<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces the concept of the prerogative and sets up a query about its nature, while the KFI provides a specific definition. The <|retrieval|> token is inserted to indicate where additional historical context might be needed to support the definition.",
"Ideal Knowledge": "A comprehensive explanation of the historical development of the royal prerogative, including its evolution from absolute monarchical power to a limited set of discretionary powers in modern constitutional monarchies.",
"Key Factual Information": "nothing else than the residue of discretionary or arbitrary authority which at any given time is legally left in the hands of the crown",
"Prefix reformulated": "What is the historical and factual nature of the prerogative?",
"Prefix-intent": "The prerogative appears to be historically and as a matter of fact",
"Verification": "The KFI directly defines the prerogative based on the historical and factual context set up in the PI."
},
{
"Explanation": "The PI sets up a query about the meaning of the term 'prerogative', and the KFI provides a concise definition. The <|retrieval|> token is inserted to indicate where additional information about the Crown's authority might be needed.",
"Ideal Knowledge": "A detailed explanation of the Crown's original authority in historical context, and how it has been reduced over time, leaving only certain powers as the royal prerogative.",
"Key Factual Information": "the remaining portion of the Crown's original authority",
"Prefix reformulated": "What does the term 'prerogative' refer to?",
"Prefix-intent": "The prerogative is the name of",
"Verification": "The KFI directly defines what the prerogative refers to, as prompted by the PI."
},
{
"Explanation": "The PI describes actions of the executive government and sets up a query about their legal basis, while the KFI provides the specific source of authority. The <|retrieval|> token is inserted to indicate where additional information about the scope of the prerogative might be needed.",
"Ideal Knowledge": "A comprehensive list of powers and actions that fall under the royal prerogative, explaining how these allow the executive to act without specific parliamentary approval in certain areas.",
"Key Factual Information": "the prerogative",
"Prefix reformulated": "What is the basis for executive government actions that don't require an Act of Parliament?",
"Prefix-intent": "Every act which the executive government can lawfully do without the authority of an Act of Parliament is done in virtue of",
"Verification": "The KFI directly answers what allows the executive government to act without parliamentary authority, as set up by the PI."
}
]
} |
While many commentators follow the Diceyan view , there are constitutional lawyers who prefer the definition given by William Blackstone in the 1760s :
| [
{
"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 While many commentators follow the Diceyan view , there are constitutional lawyers who prefer the definition given by William Blackstone in the 1760s : \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 word prerogative we usually understand that special pre @-@ eminence which the King hath , over and above all other persons , and out of the ordinary course of common law , in right of his regal dignity ... it can only be applied to those rights and capacities which the King enjoys alone , in contradiction to others , and not to those which he enjoys in common with any of his subjects .
| [
{
"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 word prerogative we usually understand that special pre @-@ eminence which the King hath , over and above all other persons , and out of the ordinary course of common law , in right of his regal dignity ... it can only be applied to those rights and capacities which the King enjoys alone , in contradiction to others , and not to those which he enjoys in common with any of his subjects . \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
} |
Dicey 's opinion that any action of governance by the monarch beyond statute is under the prerogative diverges from Blackstone 's that the prerogative simply covers those actions that no other person or body in the United Kingdom can undertake , such as the dissolution of Parliament . Case law exists to support both views . Blackstone ’ s notion of the prerogative being the powers of an exclusive nature was favoured by Lord Parmoor in the De Keyser ’ s Royal Hotel case of 1920 , but some difficulty with it was expressed by Lord Reid in the Burmah Oil case of 1965 . A clear distinction has not been necessary in the relevant cases , and the courts may never need to settle the question as few cases deal directly with the prerogative itself .
| [
{
"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 Dicey 's opinion that any action of governance by the monarch beyond statute is under the prerogative diverges from Blackstone 's that the prerogative simply covers those actions that no other person or body in the United Kingdom can undertake , such as the dissolution of Parliament . Case law exists to support both views . Blackstone ’ s notion of the prerogative being the powers of an exclusive nature was favoured by Lord Parmoor in the De Keyser ’ s Royal Hotel case of 1920 , but some difficulty with it was expressed by Lord Reid in the Burmah Oil case of 1965 . A clear distinction has not been necessary in the relevant cases , and the courts may never need to settle the question as few cases deal directly with the prerogative itself . \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 royal prerogative originated as the personal power of the monarch . From the 13th century in England , as in France , the monarch was all @-@ powerful , but this absolute power was checked by " the recrudescence of feudal turbulence in the fourteenth and fifteenth centuries " . An early attempt to define the royal prerogative was stated by Richard II 's judges in 1387 .
| [
{
"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 royal prerogative originated as the personal power of the monarch . From the 13th century in England , as in France , the monarch was all @-@ powerful , but this absolute power was checked by \" the recrudescence of feudal turbulence in the fourteenth and fifteenth centuries \" . An early attempt to define the royal prerogative was stated by Richard II 's judges in 1387 . \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
} |
During the 16th century , this " turbulence " began to recede , and the monarch became truly independent . Under Henry VIII and his successors , the king was the head of the Protestant English church , and therefore not answerable to the clergy . The rise of Parliament in this period , however , was problematic . While the monarch was " the predominant partner in the English constitution " , the courts stopped short of declaring him all @-@ powerful , recognizing the role that Parliament played . In Ferrer 's Case , Henry recognised this , noting that he was far more powerful with the consent of Parliament than without . Nowhere was this more apparent than in the matter of taxation : Sir Thomas Smith and other writers of the period pointed out the monarch could not impose taxation without Parliament 's consent .
| [
{
"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 During the 16th century , this \" turbulence \" began to recede , and the monarch became truly independent . Under Henry VIII and his successors , the king was the head of the Protestant English church , and therefore not answerable to the clergy . The rise of Parliament in this period , however , was problematic . While the monarch was \" the predominant partner in the English constitution \" , the courts stopped short of declaring him all @-@ powerful , recognizing the role that Parliament played . In Ferrer 's Case , Henry recognised this , noting that he was far more powerful with the consent of Parliament than without . Nowhere was this more apparent than in the matter of taxation : Sir Thomas Smith and other writers of the period pointed out the monarch could not impose taxation without Parliament 's consent . \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": "During the 16th century , this \" turbulence \" began to recede , and <|start-PI|> the monarch became <|end-PI|> <|retrieval|> <|start-KFI|> truly independent <|end-KFI|> . <|start-PI|> Under Henry VIII and his successors , the king was <|end-PI|> <|retrieval|> <|start-KFI|> the head of the Protestant English church , and therefore not answerable to the clergy <|end-KFI|> . The rise of Parliament in this period , however , was problematic . <|start-PI|> While the monarch was \" the predominant partner in the English constitution \" , the courts stopped short of declaring him <|end-PI|> <|retrieval|> <|start-KFI|> all @-@ powerful , recognizing the role that Parliament played <|end-KFI|> . <|start-PI|> In Ferrer 's Case , Henry recognised this , noting that he was <|end-PI|> <|retrieval|> <|start-KFI|> far more powerful with the consent of Parliament than without <|end-KFI|> . Nowhere was this more apparent than in the matter of taxation : <|start-PI|> Sir Thomas Smith and other writers of the period pointed out the monarch could not <|end-PI|> <|retrieval|> <|start-KFI|> impose taxation without Parliament 's consent <|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of the 16th century and introduces a change in the monarch's status, while the KFI provides the specific outcome. The <|retrieval|> token is inserted to indicate where additional information about the monarch's changing status might be needed.",
"Ideal Knowledge": "Information about the political and religious changes in 16th century England, focusing on the increasing power and independence of the monarchy.",
"Key Factual Information": "truly independent",
"Prefix reformulated": "How did the monarch's status change during the 16th century?",
"Prefix-intent": "During the 16th century , this \" turbulence \" began to recede , and the monarch became",
"Verification": "The KFI directly answers how the monarch's status changed, as prompted by the PI."
},
{
"Explanation": "The PI introduces the reign of Henry VIII and his successors, setting up a query about the king's role, while the KFI provides the specific position and its consequences. The <|retrieval|> token is inserted to indicate where additional information about the religious changes might be needed.",
"Ideal Knowledge": "Detailed information about the English Reformation, the establishment of the Church of England, and the monarch's role as the Supreme Head of the Church.",
"Key Factual Information": "the head of the Protestant English church , and therefore not answerable to the clergy",
"Prefix reformulated": "What was the king's role under Henry VIII and his successors?",
"Prefix-intent": "Under Henry VIII and his successors , the king was",
"Verification": "The KFI directly provides information about the king's role and its implications, as prompted by the PI."
},
{
"Explanation": "The PI sets up the monarch's position in the constitution and introduces the courts' perspective, while the KFI provides the specific limitation and recognition of Parliament. The <|retrieval|> token is inserted to indicate where additional information about the constitutional balance might be needed.",
"Ideal Knowledge": "Information about the balance of power in the English government during this period, including court decisions and the evolving role of Parliament.",
"Key Factual Information": "all @-@ powerful , recognizing the role that Parliament played",
"Prefix reformulated": "How did the courts view the monarch's power in the English constitution?",
"Prefix-intent": "While the monarch was \" the predominant partner in the English constitution \" , the courts stopped short of declaring him",
"Verification": "The KFI directly describes the courts' view of the monarch's power and acknowledges Parliament's role, as prompted by the PI."
},
{
"Explanation": "The PI introduces Ferrer's Case and Henry's recognition, setting up a query about his understanding of power, while the KFI provides the specific insight. The <|retrieval|> token is inserted to indicate where additional information about this legal case might be needed.",
"Ideal Knowledge": "Details about Ferrer's Case and its significance in defining the relationship between the monarch and Parliament, including direct quotes or paraphrases from Henry VIII.",
"Key Factual Information": "far more powerful with the consent of Parliament than without",
"Prefix reformulated": "What did Henry recognize about his power in Ferrer's Case?",
"Prefix-intent": "In Ferrer 's Case , Henry recognised this , noting that he was",
"Verification": "The KFI directly provides Henry's recognition of Parliament's importance to his power, as prompted by the PI."
},
{
"Explanation": "The PI introduces Sir Thomas Smith and other writers' observations, setting up a query about limitations on the monarch's power, while the KFI provides the specific restriction regarding taxation. The <|retrieval|> token is inserted to indicate where additional information about constitutional writings of the period might be needed.",
"Ideal Knowledge": "Information about the writings of Sir Thomas Smith and his contemporaries on the English constitution, particularly focusing on the monarch's limitations in matters of taxation.",
"Key Factual Information": "impose taxation without Parliament 's consent",
"Prefix reformulated": "What limitation on the monarch's power did Sir Thomas Smith and other writers point out?",
"Prefix-intent": "Sir Thomas Smith and other writers of the period pointed out the monarch could not",
"Verification": "The KFI directly states the limitation on the monarch's power regarding taxation, as prompted by the PI."
}
]
} |
At the same time , Henry and his descendants normally followed the will of the courts , despite the fact they were theoretically not bound by judges . William Holdsworth infers that by regularly asking the legal officers of the crown and judiciary for legal advice and consent , Henry recognised the need for a stable government to follow the law . He also contends that the view that the law is supreme over all " was the view of all the leading lawyers and statesmen and publicists of the Tudor period " . It was accepted that while the King had " unfettered discretion " , he was limited in areas where the courts had imposed conditions on the use of the prerogative , or where he had chosen to do so .
| [
{
"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 same time , Henry and his descendants normally followed the will of the courts , despite the fact they were theoretically not bound by judges . William Holdsworth infers that by regularly asking the legal officers of the crown and judiciary for legal advice and consent , Henry recognised the need for a stable government to follow the law . He also contends that the view that the law is supreme over all \" was the view of all the leading lawyers and statesmen and publicists of the Tudor period \" . It was accepted that while the King had \" unfettered discretion \" , he was limited in areas where the courts had imposed conditions on the use of the prerogative , or where he had chosen to do so . \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 dent in this stability came about in 1607 , with the Case of Prohibitions . James VI and I claimed that as monarch , he had a divine right to sit as a judge and interpret the common law as he saw fit . Led by Sir Edward Coke , the judiciary rejected this idea , stating that while the monarch was not subject to any individual , he was subject to the law . Until he had gained sufficient knowledge of the law , he had no right to interpret it ; Coke pointed out that such knowledge " demanded mastery of an artificial reason ... which requires long study and experience , before that a man can attain to the cognizance of it " . Similarly , in the Case of Proclamations in 1611 , Coke held that the monarch could only exercise those prerogatives he already had , and not create new ones .
| [
{
"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 dent in this stability came about in 1607 , with the Case of Prohibitions . James VI and I claimed that as monarch , he had a divine right to sit as a judge and interpret the common law as he saw fit . Led by Sir Edward Coke , the judiciary rejected this idea , stating that while the monarch was not subject to any individual , he was subject to the law . Until he had gained sufficient knowledge of the law , he had no right to interpret it ; Coke pointed out that such knowledge \" demanded mastery of an artificial reason ... which requires long study and experience , before that a man can attain to the cognizance of it \" . Similarly , in the Case of Proclamations in 1611 , Coke held that the monarch could only exercise those prerogatives he already had , and not create new ones . \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
} |
With the Glorious Revolution , King James VII and II was replaced by Queen Mary II and her husband King William III . At the same time the Bill of Rights 1689 was drafted , which cemented the monarch 's subservience to Parliament . It specifically limited the royal prerogative , with Article 1 holding that the " power of suspending the laws or the execution of laws by regal authority without consent of Parliament is illegal " , and article 4 confirming that " levying money for or to the use of the Crown by pretence of prerogative , without grant of Parliament , for longer time , or in other manner than the same is or shall be granted , is illegal " . The Bill also confirmed that Parliament had the right to limit the use of remaining prerogatives , as evidenced by the Triennial Act 1694 , which required the monarch to dismiss and call Parliament at certain times .
| [
{
"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 With the Glorious Revolution , King James VII and II was replaced by Queen Mary II and her husband King William III . At the same time the Bill of Rights 1689 was drafted , which cemented the monarch 's subservience to Parliament . It specifically limited the royal prerogative , with Article 1 holding that the \" power of suspending the laws or the execution of laws by regal authority without consent of Parliament is illegal \" , and article 4 confirming that \" levying money for or to the use of the Crown by pretence of prerogative , without grant of Parliament , for longer time , or in other manner than the same is or shall be granted , is illegal \" . The Bill also confirmed that Parliament had the right to limit the use of remaining prerogatives , as evidenced by the Triennial Act 1694 , which required the monarch to dismiss and call Parliament at certain times . \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
} |
One of the monarch 's historic prerogatives was the dissolution of Parliament , which was " perhaps the most important residual prerogative exercised personally by the sovereign , and represents the greatest potential for controversy . " This prerogative was normally exercised at the request of Parliament and the prime minister , either at his or her discretion or following a motion of no confidence . Constitutional theorists have had differing views as to whether a unilateral dissolution of Parliament would be possible today ; Sir Ivor Jennings wrote that a dissolution involves " the acquiescence of ministers " , and as such the monarch could not dissolve Parliament without ministerial consent ; " if ministers refuse to give such advice , she can do no more than dismiss them " . A. V. Dicey , however , believed that in certain extreme circumstances the monarch could dissolve Parliament single @-@ handedly , on the condition that " an occasion has arisen on which there is fair reason to suppose that the opinion of the House is not the opinion of the electors ... A dissolution is allowable , or necessary , whenever the wishes of the legislature are , or may fairly be presumed to be , different from the wishes of the nation . "
| [
{
"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 One of the monarch 's historic prerogatives was the dissolution of Parliament , which was \" perhaps the most important residual prerogative exercised personally by the sovereign , and represents the greatest potential for controversy . \" This prerogative was normally exercised at the request of Parliament and the prime minister , either at his or her discretion or following a motion of no confidence . Constitutional theorists have had differing views as to whether a unilateral dissolution of Parliament would be possible today ; Sir Ivor Jennings wrote that a dissolution involves \" the acquiescence of ministers \" , and as such the monarch could not dissolve Parliament without ministerial consent ; \" if ministers refuse to give such advice , she can do no more than dismiss them \" . A. V. Dicey , however , believed that in certain extreme circumstances the monarch could dissolve Parliament single @-@ handedly , on the condition that \" an occasion has arisen on which there is fair reason to suppose that the opinion of the House is not the opinion of the electors ... A dissolution is allowable , or necessary , whenever the wishes of the legislature are , or may fairly be presumed to be , different from the wishes of the nation . \" \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 monarch could force the dissolution of Parliament through a refusal of royal assent ; this would inevitably lead to a government resigning . By convention , the monarch always assents to bills ; the last time the royal assent was not given was in 1704 during the reign of Queen Anne . This does not mean that the right to refuse has died : George V believed he could veto the Third Irish Home Rule Bill ; Jennings writes that " it was assumed by the King throughout that he had not only the legal power but the constitutional right to refuse assent " . The royal prerogative to dissolve Parliament was abrogated by the Fixed @-@ term Parliaments Act 2011 . Section 6 ( 1 ) of the Act however specifically states that the monarch 's power to prorogue Parliament is not affected by the Act .
| [
{
"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 monarch could force the dissolution of Parliament through a refusal of royal assent ; this would inevitably lead to a government resigning . By convention , the monarch always assents to bills ; the last time the royal assent was not given was in 1704 during the reign of Queen Anne . This does not mean that the right to refuse has died : George V believed he could veto the Third Irish Home Rule Bill ; Jennings writes that \" it was assumed by the King throughout that he had not only the legal power but the constitutional right to refuse assent \" . The royal prerogative to dissolve Parliament was abrogated by the Fixed @-@ term Parliaments Act 2011 . Section 6 ( 1 ) of the Act however specifically states that the monarch 's power to prorogue Parliament is not affected by the Act . \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 appointment of the prime minister is also , theoretically , governed by the royal prerogative . Technically the monarch may appoint as prime minister anyone she wants to appoint , but in practice the appointee is always the person who commands a majority in the House of Commons . Usually , this is the leader of the political party that is returned to Parliament with a majority of seats after a general election . Difficulties may result with a so @-@ called hung parliament , in which no party commands majority support , as last occurred in 2010 . In this situation , constitutional convention is that the previous incumbent has the first right to form a coalition government and seek appointment . If the prime minister decides to retire in the middle of a parliamentary session , as Anthony Eden did in 1957 , the monarch has no discretion . There is usually a " prime minister @-@ in @-@ waiting " who commands the support of the majority of the Commons ; he or she will near @-@ automatically be appointed .
| [
{
"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 appointment of the prime minister is also , theoretically , governed by the royal prerogative . Technically the monarch may appoint as prime minister anyone she wants to appoint , but in practice the appointee is always the person who commands a majority in the House of Commons . Usually , this is the leader of the political party that is returned to Parliament with a majority of seats after a general election . Difficulties may result with a so @-@ called hung parliament , in which no party commands majority support , as last occurred in 2010 . In this situation , constitutional convention is that the previous incumbent has the first right to form a coalition government and seek appointment . If the prime minister decides to retire in the middle of a parliamentary session , as Anthony Eden did in 1957 , the monarch has no discretion . There is usually a \" prime minister @-@ in @-@ waiting \" who commands the support of the majority of the Commons ; he or she will near @-@ automatically be appointed . \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 appointment of the prime minister is also , theoretically , governed by<|end-PI|> <|retrieval|> <|start-KFI|>the royal prerogative<|end-KFI|> . <|start-PI|>Technically the monarch may appoint as prime minister<|end-PI|> <|retrieval|> <|start-KFI|>anyone she wants to appoint , but in practice the appointee is always the person who commands a majority in the House of Commons<|end-KFI|> . <|start-PI|>Usually , this is the leader of the political party that is returned to Parliament with<|end-PI|> <|retrieval|> <|start-KFI|>a majority of seats after a general election<|end-KFI|> . <|start-PI|>Difficulties may result with a so @-@ called hung parliament , in which<|end-PI|> <|retrieval|> <|start-KFI|>no party commands majority support , as last occurred in 2010<|end-KFI|> . <|start-PI|>In this situation , constitutional convention is that<|end-PI|> <|retrieval|> <|start-KFI|>the previous incumbent has the first right to form a coalition government and seek appointment<|end-KFI|> . <|start-PI|>If the prime minister decides to retire in the middle of a parliamentary session , as Anthony Eden did in 1957 ,<|end-PI|> <|retrieval|> <|start-KFI|>the monarch has no discretion<|end-KFI|> . There is usually a \" prime minister @-@ in @-@ waiting \" who commands the support of the majority of the Commons ; he or she will near @-@ automatically be appointed .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the topic of prime minister appointment, and the KFI provides the specific governing principle. The <|retrieval|> token is inserted to indicate where additional information about the royal prerogative might be needed.",
"Ideal Knowledge": "Information on the UK's constitutional monarchy, explaining the role of royal prerogative in government appointments, particularly for the position of prime minister.",
"Key Factual Information": "the royal prerogative",
"Prefix reformulated": "What governs the appointment of the prime minister in the UK?",
"Prefix-intent": "The appointment of the prime minister is also , theoretically , governed by",
"Verification": "The KFI directly states what governs the appointment of the prime minister, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the monarch's power in appointing a prime minister, and the KFI provides both the theoretical and practical answers. The <|retrieval|> token is inserted to indicate where additional information about the UK's political system might be needed.",
"Ideal Knowledge": "Detailed explanation of the UK's parliamentary system, including the relationship between the monarch and the House of Commons in forming a government.",
"Key Factual Information": "anyone she wants to appoint , but in practice the appointee is always the person who commands a majority in the House of Commons",
"Prefix reformulated": "Who can the monarch appoint as prime minister in the UK?",
"Prefix-intent": "Technically the monarch may appoint as prime minister",
"Verification": "The KFI directly answers who the monarch can appoint and explains the practical reality, as prompted by the PI."
},
{
"Explanation": "The PI introduces the usual scenario for prime minister appointment, and the KFI provides the specific condition. The <|retrieval|> token is inserted to indicate where additional information about UK elections might be needed.",
"Ideal Knowledge": "Information on the UK's electoral system, explaining how parliamentary seats are allocated and what constitutes a majority government.",
"Key Factual Information": "a majority of seats after a general election",
"Prefix reformulated": "What type of majority does the leader of the political party usually need to become prime minister?",
"Prefix-intent": "Usually , this is the leader of the political party that is returned to Parliament with",
"Verification": "The KFI directly states the type of majority needed, as prompted by the PI."
},
{
"Explanation": "The PI introduces the concept of a hung parliament, and the KFI provides its definition and a recent example. The <|retrieval|> token is inserted to indicate where additional information about hung parliaments might be needed.",
"Ideal Knowledge": "Detailed explanation of hung parliaments in the UK, including historical instances and their impact on government formation.",
"Key Factual Information": "no party commands majority support , as last occurred in 2010",
"Prefix reformulated": "What is a hung parliament in the UK?",
"Prefix-intent": "Difficulties may result with a so @-@ called hung parliament , in which",
"Verification": "The KFI directly defines a hung parliament and provides a recent example, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the protocol in a hung parliament, and the KFI provides the specific convention. The <|retrieval|> token is inserted to indicate where additional information about UK constitutional conventions might be needed.",
"Ideal Knowledge": "Information on UK constitutional conventions, particularly those related to government formation in complex political situations like hung parliaments.",
"Key Factual Information": "the previous incumbent has the first right to form a coalition government and seek appointment",
"Prefix reformulated": "What is the constitutional convention in a hung parliament situation in the UK?",
"Prefix-intent": "In this situation , constitutional convention is that",
"Verification": "The KFI directly states the constitutional convention for hung parliaments, as prompted by the PI."
},
{
"Explanation": "The PI introduces a scenario of mid-term retirement, and the KFI provides the constitutional response. The <|retrieval|> token is inserted to indicate where additional information about mid-term changes in UK government might be needed.",
"Ideal Knowledge": "Historical information about mid-term prime ministerial changes in the UK, including specific cases like Anthony Eden's resignation and the constitutional procedures followed.",
"Key Factual Information": "the monarch has no discretion",
"Prefix reformulated": "What happens if the UK prime minister retires mid-session, like Anthony Eden in 1957?",
"Prefix-intent": "If the prime minister decides to retire in the middle of a parliamentary session , as Anthony Eden did in 1957 ,",
"Verification": "The KFI directly states the monarch's role in this situation, as prompted by the PI."
}
]
} |
The most noted prerogative power that affects the judicial system is the prerogative of mercy , which has two elements : the granting of pardons and the granting of nolle prosequi . Pardons may eliminate the " pains , penalties and punishments " from a criminal conviction , though they do not remove convictions themselves . This power is commonly exercised on the advice of the Secretary of State for the Home Department ; the monarch has no direct involvement in its use . Exercises of this power may also take the form of commutations , a limited form of pardon where the sentences is reduced , on certain conditions . The granting of a pardon is not subject to judicial review , as confirmed by Council of Civil Service Unions v Minister for the Civil Service , but the courts have chosen to criticise its application or lack thereof , as in R v Secretary of State for the Home Department , ex parte Bentley . Granting nolle prosequi is done by the Attorney General of England and Wales ( or the equivalent in Scotland or Northern Ireland ) in the name of the crown , to stop legal proceedings against an individual . This is not reviewable by the courts , as confirmed by R v Comptroller of Patents , and does not count as an acquittal ; the defendant may be brought before the courts on the same charge at a later date .
| [
{
"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 most noted prerogative power that affects the judicial system is the prerogative of mercy , which has two elements : the granting of pardons and the granting of nolle prosequi . Pardons may eliminate the \" pains , penalties and punishments \" from a criminal conviction , though they do not remove convictions themselves . This power is commonly exercised on the advice of the Secretary of State for the Home Department ; the monarch has no direct involvement in its use . Exercises of this power may also take the form of commutations , a limited form of pardon where the sentences is reduced , on certain conditions . The granting of a pardon is not subject to judicial review , as confirmed by Council of Civil Service Unions v Minister for the Civil Service , but the courts have chosen to criticise its application or lack thereof , as in R v Secretary of State for the Home Department , ex parte Bentley . Granting nolle prosequi is done by the Attorney General of England and Wales ( or the equivalent in Scotland or Northern Ireland ) in the name of the crown , to stop legal proceedings against an individual . This is not reviewable by the courts , as confirmed by R v Comptroller of Patents , and does not count as an acquittal ; the defendant may be brought before the courts on the same charge at a later date . \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 royal prerogative is in much use in the realm of foreign affairs . It is the monarch who recognises foreign states ( although several statutes regulate the immunities enjoyed by their heads and diplomatic representatives ) , issues declarations of war and peace , and forms international treaties . The monarch also has the power to annex territory , as was done in 1955 with the island of Rockall . Once territory has been annexed , the monarch has complete discretion as to the extent to which the government will take over the former government 's liabilities ; this was confirmed in West Rand Central Gold Mining Company v The King . The monarch also has the power to alter British territorial waters and cede territory . Her freedom to do these things in practice is doubtful , in that they might deprive British citizens of their nationality and rights . When the island of Heligoland was ceded to Germany in 1890 , Parliamentary approval was first sought . The monarch can also regulate colonies and dependent territories by exercising the prerogative through Orders in Council . The courts have long fought against the monarch 's use of this power : in R ( Bancoult ) v Secretary of State for Foreign and Commonwealth Affairs ( No 2 ) , the Court of Appeal ruled that using Orders @-@ in @-@ Council to frustrate judicial rulings was an unlawful abuse of power , although this ruling was later overturned .
| [
{
"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 royal prerogative is in much use in the realm of foreign affairs . It is the monarch who recognises foreign states ( although several statutes regulate the immunities enjoyed by their heads and diplomatic representatives ) , issues declarations of war and peace , and forms international treaties . The monarch also has the power to annex territory , as was done in 1955 with the island of Rockall . Once territory has been annexed , the monarch has complete discretion as to the extent to which the government will take over the former government 's liabilities ; this was confirmed in West Rand Central Gold Mining Company v The King . The monarch also has the power to alter British territorial waters and cede territory . Her freedom to do these things in practice is doubtful , in that they might deprive British citizens of their nationality and rights . When the island of Heligoland was ceded to Germany in 1890 , Parliamentary approval was first sought . The monarch can also regulate colonies and dependent territories by exercising the prerogative through Orders in Council . The courts have long fought against the monarch 's use of this power : in R ( Bancoult ) v Secretary of State for Foreign and Commonwealth Affairs ( No 2 ) , the Court of Appeal ruled that using Orders @-@ in @-@ Council to frustrate judicial rulings was an unlawful abuse of power , although this ruling was later overturned . \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
} |
British passports are also issued under the prerogative , though these are also covered by statute law . Under the common law , citizens have the right freely to leave and enter the United Kingdom . In R v Foreign Secretary ex parte Everett , the courts held that it was their right to review the granting of passports to , and the withholding of passports from , British citizens . The writ of ne exeat regno is also used to prevent a person leaving the country . The right to make treaties is a disputed prerogative power : under Blackstone 's definition , a prerogative power must be one unique to the monarch .
| [
{
"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 British passports are also issued under the prerogative , though these are also covered by statute law . Under the common law , citizens have the right freely to leave and enter the United Kingdom . In R v Foreign Secretary ex parte Everett , the courts held that it was their right to review the granting of passports to , and the withholding of passports from , British citizens . The writ of ne exeat regno is also used to prevent a person leaving the country . The right to make treaties is a disputed prerogative power : under Blackstone 's definition , a prerogative power must be one unique to the monarch . \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 monarch also has power to exercise her prerogative over the granting of honours , the regulation of the armed forces and ecclesiastical appointments . Although the granting of most honours is normally decided by the executive , the monarch is still the person who technically awards them . Exceptions to this rule are membership of the Order of the Garter , the Order of the Thistle , the Order of Merit , the Royal Victorian Order and the Royal Victorian Chain , which the monarch has complete discretion to grant . In relation to the armed forces , the monarch is the Commander in Chief , and members are regulated under the royal prerogative . Most statutes do not apply to the armed forces , although some areas , such as military discipline , are governed by Acts of Parliament . Under the Crown Proceedings Act 1947 , the monarch is the sole authority for the armed forces , and as such their organisation , disposition and control cannot be questioned by the courts . This exercise of prerogative power gives the Crown authority to recruit members of the armed forces , appoint commissioned officers , and establish agreements with foreign governments to station troops in their territory . The prerogative empowers the monarch to appoint bishops and archbishops in the Church of England , and to regulate the printing and licensing of the authorised Church of England version of the Bible .
| [
{
"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 monarch also has power to exercise her prerogative over the granting of honours , the regulation of the armed forces and ecclesiastical appointments . Although the granting of most honours is normally decided by the executive , the monarch is still the person who technically awards them . Exceptions to this rule are membership of the Order of the Garter , the Order of the Thistle , the Order of Merit , the Royal Victorian Order and the Royal Victorian Chain , which the monarch has complete discretion to grant . In relation to the armed forces , the monarch is the Commander in Chief , and members are regulated under the royal prerogative . Most statutes do not apply to the armed forces , although some areas , such as military discipline , are governed by Acts of Parliament . Under the Crown Proceedings Act 1947 , the monarch is the sole authority for the armed forces , and as such their organisation , disposition and control cannot be questioned by the courts . This exercise of prerogative power gives the Crown authority to recruit members of the armed forces , appoint commissioned officers , and establish agreements with foreign governments to station troops in their territory . The prerogative empowers the monarch to appoint bishops and archbishops in the Church of England , and to regulate the printing and licensing of the authorised Church of England version of the Bible . \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
} |
R v Secretary of State for the Home Department , ex parte Northumbria Police Authority , recognised that the prerogative also includes the power to " take all reasonable steps to preserve the Queen 's peace " , and in Burmah Oil Co. v Lord Advocate , the House of Lords took the view that it extended to " doing all those things in an emergency which are necessary for the conduct of [ the Second World War ] . "
| [
{
"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 R v Secretary of State for the Home Department , ex parte Northumbria Police Authority , recognised that the prerogative also includes the power to \" take all reasonable steps to preserve the Queen 's peace \" , and in Burmah Oil Co. v Lord Advocate , the House of Lords took the view that it extended to \" doing all those things in an emergency which are necessary for the conduct of [ the Second World War ] . \" \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
} |
Today , the monarch exercises the prerogative almost exclusively in line with the advice of her government . Leyland notes 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 Today , the monarch exercises the prerogative almost exclusively in line with the advice of her government . Leyland notes 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": null,
"none": "",
"results": null
} |
The present Queen ... is kept very closely in touch with the exercise of governmental power by means of a weekly audience with the prime minister during which she is fully briefed about the affairs of government ... [ But it ] should be emphasised that the prime minister is not under any obligation to take account of royal opinions .
| [
{
"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 present Queen ... is kept very closely in touch with the exercise of governmental power by means of a weekly audience with the prime minister during which she is fully briefed about the affairs of government ... [ But it ] should be emphasised that the prime minister is not under any obligation to take account of royal opinions . \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 simple terms , the prerogative is used to govern the realm in the name of the Crown ; although the monarch has the " right to be consulted , the right to encourage , and the right to warn " , her role involves no exercise of discretion .
| [
{
"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 simple terms , the prerogative is used to govern the realm in the name of the Crown ; although the monarch has the \" right to be consulted , the right to encourage , and the right to warn \" , her role involves no exercise of discretion . \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
} |
Today , some prerogative powers are directly exercised by ministers without the approval of Parliament , including the powers of declaring war and of making peace , the issue of passports , and the granting of honours . Prerogative powers are exercised nominally by the monarch , but on the advice of the prime minister ( whom the monarch meets weekly ) and of the cabinet . Some key functions of the British government are still executed by virtue of the royal prerogative , but generally the usage of the prerogative has been diminishing as functions are progressively put on a statutory basis .
| [
{
"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 Today , some prerogative powers are directly exercised by ministers without the approval of Parliament , including the powers of declaring war and of making peace , the issue of passports , and the granting of honours . Prerogative powers are exercised nominally by the monarch , but on the advice of the prime minister ( whom the monarch meets weekly ) and of the cabinet . Some key functions of the British government are still executed by virtue of the royal prerogative , but generally the usage of the prerogative has been diminishing as functions are progressively put on a statutory basis . \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
} |
Several influential decisions of the House of Lords have determined the limited scope for the use of prerogative powers . In 1915 , an appeal was made to the House of Lords , Re Petition of Right ( ' Shoreham Aerodrome Case ' ) , but during the appeal the case was settled and the appeal withdrawn when the Crown agreed to pay compensation . The appeal was from a unanimous decision of the Court of Appeal that the Crown , both under the statutory Defence of the Realm Regulations and by the royal prerogative , was entitled to take and occupy , for military purposes in wartime , a commercial airfield on the south coast . The government argued that this action was to defend against an invasion ; the courts held that for the prerogative to be exercised , the government must demonstrate that a threat of invasion exists . This was backed up by The Zamora ( 1916 ) , where the Privy Council , on appeal from the Prize Court , held generally that to exercise a power not granted by statute ( such as a prerogative power ) the government must prove to the court that the exercise is justified . The next decision came in Attorney General v De Keyser 's Royal Hotel Ltd ( 1920 ) , where the House of Lords confirmed that a statutory provision in an area where prerogative powers are in use " abridges the Royal Prerogative while it is in force to this extent – that the Crown can only do the particular thing under and in accordance with the statutory provisions , and that its prerogative power to do that thing is in abeyance " .
| [
{
"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 Several influential decisions of the House of Lords have determined the limited scope for the use of prerogative powers . In 1915 , an appeal was made to the House of Lords , Re Petition of Right ( ' Shoreham Aerodrome Case ' ) , but during the appeal the case was settled and the appeal withdrawn when the Crown agreed to pay compensation . The appeal was from a unanimous decision of the Court of Appeal that the Crown , both under the statutory Defence of the Realm Regulations and by the royal prerogative , was entitled to take and occupy , for military purposes in wartime , a commercial airfield on the south coast . The government argued that this action was to defend against an invasion ; the courts held that for the prerogative to be exercised , the government must demonstrate that a threat of invasion exists . This was backed up by The Zamora ( 1916 ) , where the Privy Council , on appeal from the Prize Court , held generally that to exercise a power not granted by statute ( such as a prerogative power ) the government must prove to the court that the exercise is justified . The next decision came in Attorney General v De Keyser 's Royal Hotel Ltd ( 1920 ) , where the House of Lords confirmed that a statutory provision in an area where prerogative powers are in use \" abridges the Royal Prerogative while it is in force to this extent – that the Crown can only do the particular thing under and in accordance with the statutory provisions , and that its prerogative power to do that thing is in abeyance \" . \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
} |
This principle of statutory superiority was extended in Laker Airway Ltd v Department of Trade , concerning the revocation of a commercial airline operator 's licence ( December 1976 ) , where it was confirmed that prerogative powers could not be used to contradict a statutory provision , and that in situations to which the power and the statute both applied , the power could only be used to further the aim of the statute . Another extension came with R v Secretary of State for the Home Department , ex parte Fire Brigades Union , where the Court of Appeal held that even if a statute had not yet come into force , the prerogative could not be used to alter this statute to " conflict with Parliament 's wishes " .
| [
{
"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 This principle of statutory superiority was extended in Laker Airway Ltd v Department of Trade , concerning the revocation of a commercial airline operator 's licence ( December 1976 ) , where it was confirmed that prerogative powers could not be used to contradict a statutory provision , and that in situations to which the power and the statute both applied , the power could only be used to further the aim of the statute . Another extension came with R v Secretary of State for the Home Department , ex parte Fire Brigades Union , where the Court of Appeal held that even if a statute had not yet come into force , the prerogative could not be used to alter this statute to \" conflict with Parliament 's wishes \" . \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
} |
Before the modern judicial review procedure superseded the petition of right as the remedy for challenging the validity of a prerogative power , the courts were traditionally only willing to state whether or not powers existed , not whether they had been used appropriately . They therefore applied only the first of the Wednesbury tests : whether the use was illegal . Constitutional scholars such as William Blackstone consider this appropriate :
| [
{
"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 Before the modern judicial review procedure superseded the petition of right as the remedy for challenging the validity of a prerogative power , the courts were traditionally only willing to state whether or not powers existed , not whether they had been used appropriately . They therefore applied only the first of the Wednesbury tests : whether the use was illegal . Constitutional scholars such as William Blackstone consider this appropriate : \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 the exertion therefore of those prerogatives , which the law has given him , the King is irresistible and absolute , according to the forms of the constitution . And yet if the consequence of that exertion be manifestly to the grievance or dishonour of the kingdom , the Parliament will call his advisers to a just and severe account .
| [
{
"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 the exertion therefore of those prerogatives , which the law has given him , the King is irresistible and absolute , according to the forms of the constitution . And yet if the consequence of that exertion be manifestly to the grievance or dishonour of the kingdom , the Parliament will call his advisers to a just and severe account . \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
} |
During the 1960s and 70s this attitude was changing , with Lord Denning saying in the Laker Airway case that " seeing that the prerogative is a discretionary power to be exercised for the public good , it follows that its exercise can be examined by the courts just as any other discretionary power which is vested in the executive . " The most authoritative case on the matter is Council of Civil Service Unions v Minister for the Civil Service , generally known as the GCHQ case . The House of Lords confirmed that the application of judicial review would be dependent on the nature of the government 's powers , not their source . Foreign policy and national security powers are considered outside the scope of judicial review , while the prerogative of mercy is considered within it , as per R v Secretary of State for the Home Department , ex parte Bentley .
| [
{
"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 During the 1960s and 70s this attitude was changing , with Lord Denning saying in the Laker Airway case that \" seeing that the prerogative is a discretionary power to be exercised for the public good , it follows that its exercise can be examined by the courts just as any other discretionary power which is vested in the executive . \" The most authoritative case on the matter is Council of Civil Service Unions v Minister for the Civil Service , generally known as the GCHQ case . The House of Lords confirmed that the application of judicial review would be dependent on the nature of the government 's powers , not their source . Foreign policy and national security powers are considered outside the scope of judicial review , while the prerogative of mercy is considered within it , as per R v Secretary of State for the Home Department , ex parte Bentley . \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
} |
Abolition of the royal prerogative is not on the immediate horizon , and recent movements to abolish the role of the monarchy and its royal prerogative in government have been unsuccessful . The Ministry of Justice undertook a " review of executive Royal Prerogative powers " in October 2009 . Former Labour MP and cabinet minister Tony Benn campaigned unsuccessfully for the abolition of the royal prerogative in the United Kingdom in the 1990s , arguing that all governmental powers in effect exercised on the advice of the prime minister and cabinet should be subject to parliamentary scrutiny and require parliamentary approval . Later governments argued that such is the breadth of topics covered by the royal prerogative that requiring parliamentary approval in each instance where the prerogative is currently used would overwhelm parliamentary time and slow the enactment of legislation .
| [
{
"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 Abolition of the royal prerogative is not on the immediate horizon , and recent movements to abolish the role of the monarchy and its royal prerogative in government have been unsuccessful . The Ministry of Justice undertook a \" review of executive Royal Prerogative powers \" in October 2009 . Former Labour MP and cabinet minister Tony Benn campaigned unsuccessfully for the abolition of the royal prerogative in the United Kingdom in the 1990s , arguing that all governmental powers in effect exercised on the advice of the prime minister and cabinet should be subject to parliamentary scrutiny and require parliamentary approval . Later governments argued that such is the breadth of topics covered by the royal prerogative that requiring parliamentary approval in each instance where the prerogative is currently used would overwhelm parliamentary time and slow the enactment of legislation . \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
} |
Mount Jackson ( Mount Andrew Jackson and Mount Ernest Gruening ) is a mountain that dominates the upland of the southern part of the Antarctic Peninsula . It is located in Palmer Land , within the Antarctic claims of Argentina , Chile and the United Kingdom . With an elevation of 3 @,@ 184 metres ( 10 @,@ 446 ft ) , Mount Jackson is the highest mountain in the Antarctic Peninsula and the British Antarctic Territory . Discovered by members of the United States Antarctic Service , 1939 – 41 , it was named for Andrew Jackson , the seventh President of the United States . The first ascent of Mount Jackson was made by a team led by John Crabbe Cunningham of the British Antarctic Survey ( BAS ) in 1964 . Mount Jackson 's geology was studied in 1972 as part of the Palmer Island investigations by a team of geologists .
| [
{
"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 Mount Jackson ( Mount Andrew Jackson and Mount Ernest Gruening ) is a mountain that dominates the upland of the southern part of the Antarctic Peninsula . It is located in Palmer Land , within the Antarctic claims of Argentina , Chile and the United Kingdom . With an elevation of 3 @,@ 184 metres ( 10 @,@ 446 ft ) , Mount Jackson is the highest mountain in the Antarctic Peninsula and the British Antarctic Territory . Discovered by members of the United States Antarctic Service , 1939 – 41 , it was named for Andrew Jackson , the seventh President of the United States . The first ascent of Mount Jackson was made by a team led by John Crabbe Cunningham of the British Antarctic Survey ( BAS ) in 1964 . Mount Jackson 's geology was studied in 1972 as part of the Palmer Island investigations by a team of geologists . \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
} |
Mount Jackson and the Welch Mountains demarcate the central Black Coast , which is dissected by many inlets and is bounded on the west by Dyer Plateau of central Palmer Land . The two mountains rise above the ice shelf with reliefs of about 1 @,@ 200 – 1 @,@ 500 metres ( 3 @,@ 900 – 4 @,@ 900 ft ) towards the east . They are interconnected by icefall zones forming a plateau with steep snow slopes . Mount Jackson rises from its southeast flanks , displaying a steeple summit , while the north flank is occupied by a vast cirque . Supraglacial moraines on the mountain 's east side measure between 1 – 5 kilometres ( 0 @.@ 62 – 3 @.@ 11 mi ) in length and display boulders at their distal ends .
| [
{
"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 Mount Jackson and the Welch Mountains demarcate the central Black Coast , which is dissected by many inlets and is bounded on the west by Dyer Plateau of central Palmer Land . The two mountains rise above the ice shelf with reliefs of about 1 @,@ 200 – 1 @,@ 500 metres ( 3 @,@ 900 – 4 @,@ 900 ft ) towards the east . They are interconnected by icefall zones forming a plateau with steep snow slopes . Mount Jackson rises from its southeast flanks , displaying a steeple summit , while the north flank is occupied by a vast cirque . Supraglacial moraines on the mountain 's east side measure between 1 – 5 kilometres ( 0 @.@ 62 – 3 @.@ 11 mi ) in length and display boulders at their distal ends . \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|>Mount Jackson and the Welch Mountains demarcate the central Black Coast , which is dissected by many inlets and is bounded on the west by<|end-PI|> <|retrieval|> <|start-KFI|>Dyer Plateau of central Palmer Land<|end-KFI|> . <|start-PI|>The two mountains rise above the ice shelf with reliefs of about<|end-PI|> <|retrieval|> <|start-KFI|>1 @,@ 200 – 1 @,@ 500 metres ( 3 @,@ 900 – 4 @,@ 900 ft ) towards the east<|end-KFI|> . They are interconnected by icefall zones forming a plateau with steep snow slopes . <|start-PI|>Mount Jackson rises from its southeast flanks , displaying<|end-PI|> <|retrieval|> <|start-KFI|>a steeple summit , while the north flank is occupied by a vast cirque<|end-KFI|> . <|start-PI|>Supraglacial moraines on the mountain 's east side measure between<|end-PI|> <|retrieval|> <|start-KFI|>1 – 5 kilometres ( 0 @.@ 62 – 3 @.@ 11 mi ) in length and display boulders at their distal ends<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the location and sets up a query about its western boundary, while the KFI provides the specific geographical feature. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "A detailed geographical description of the Black Coast region, including its boundaries and major landmarks, specifically mentioning the Dyer Plateau and its location in relation to the central Black Coast.",
"Key Factual Information": "Dyer Plateau of central Palmer Land",
"Prefix reformulated": "What bounds the central Black Coast on the west?",
"Prefix-intent": "Mount Jackson and the Welch Mountains demarcate the central Black Coast , which is dissected by many inlets and is bounded on the west by",
"Verification": "The KFI directly provides the information about what bounds the central Black Coast on the west, as requested in the PI."
},
{
"Explanation": "The PI sets up a query about the relief of the mountains, and the KFI provides the specific measurements. The <|retrieval|> token is inserted to indicate where additional information about the mountains' elevation might be needed.",
"Ideal Knowledge": "Detailed topographical information about Mount Jackson and the Welch Mountains, including their elevation above the surrounding ice shelf and how this elevation changes from west to east.",
"Key Factual Information": "1 @,@ 200 – 1 @,@ 500 metres ( 3 @,@ 900 – 4 @,@ 900 ft ) towards the east",
"Prefix reformulated": "What is the relief of Mount Jackson and the Welch Mountains above the ice shelf?",
"Prefix-intent": "The two mountains rise above the ice shelf with reliefs of about",
"Verification": "The KFI directly provides the height information requested in the PI."
},
{
"Explanation": "The PI introduces Mount Jackson and sets up a query about its features, while the KFI provides specific details about its summit and northern flank. The <|retrieval|> token is inserted to indicate where additional information about the mountain's features might be needed.",
"Ideal Knowledge": "Detailed geological and topographical information about Mount Jackson, including descriptions of its summit shape and the presence of cirques on its flanks.",
"Key Factual Information": "a steeple summit , while the north flank is occupied by a vast cirque",
"Prefix reformulated": "What does Mount Jackson display as it rises from its southeast flanks?",
"Prefix-intent": "Mount Jackson rises from its southeast flanks , displaying",
"Verification": "The KFI directly describes the features of Mount Jackson as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the size of the supraglacial moraines, and the KFI provides the specific measurements and additional details. The <|retrieval|> token is inserted to indicate where additional information about moraines might be needed.",
"Ideal Knowledge": "Detailed information about supraglacial moraines, their typical sizes, and characteristics, specifically focusing on those found on Mount Jackson's east side.",
"Key Factual Information": "1 – 5 kilometres ( 0 @.@ 62 – 3 @.@ 11 mi ) in length and display boulders at their distal ends",
"Prefix reformulated": "What is the length of the supraglacial moraines on Mount Jackson's east side?",
"Prefix-intent": "Supraglacial moraines on the mountain 's east side measure between",
"Verification": "The KFI directly provides the length information requested in the PI and additional details about the moraines."
}
]
} |
The first topographic mapping of Mount Jackson was carried out in November 1940 by a sledge party of the United States Antarctic Service ( USAS ) . The ground survey was facilitated by aerial photographs and aerial observations . The height was estimated at 4 @,@ 200 metres ( 13 @,@ 800 ft ) , and the mountain was named Mount Ernest Gruening after the Governor of the Alaska Territory at that time , Ernest Gruening . USAS later renamed it Mount Jackson for the seventh President of the United States . President Jackson signed the bill into law authorizing the United States Exploring Expedition of 1838 – 42 . The expedition , led by Lt. Charles Wilkes , included exploration and surveying of the Pacific Ocean and surrounding lands , including Antarctica .
| [
{
"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 topographic mapping of Mount Jackson was carried out in November 1940 by a sledge party of the United States Antarctic Service ( USAS ) . The ground survey was facilitated by aerial photographs and aerial observations . The height was estimated at 4 @,@ 200 metres ( 13 @,@ 800 ft ) , and the mountain was named Mount Ernest Gruening after the Governor of the Alaska Territory at that time , Ernest Gruening . USAS later renamed it Mount Jackson for the seventh President of the United States . President Jackson signed the bill into law authorizing the United States Exploring Expedition of 1838 – 42 . The expedition , led by Lt. Charles Wilkes , included exploration and surveying of the Pacific Ocean and surrounding lands , including Antarctica . \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 first topographic mapping of Mount Jackson was carried out in<|end-PI|> <|retrieval|> <|start-KFI|>November 1940 by a sledge party of the United States Antarctic Service ( USAS )<|end-KFI|> . The ground survey was facilitated by aerial photographs and aerial observations . <|start-PI|>The height was estimated at<|end-PI|> <|retrieval|> <|start-KFI|>4 @,@ 200 metres ( 13 @,@ 800 ft )<|end-KFI|> , and <|start-PI|>the mountain was named Mount Ernest Gruening after<|end-PI|> <|retrieval|> <|start-KFI|>the Governor of the Alaska Territory at that time , Ernest Gruening<|end-KFI|> . <|start-PI|>USAS later renamed it Mount Jackson for<|end-PI|> <|retrieval|> <|start-KFI|>the seventh President of the United States<|end-KFI|> . <|start-PI|>President Jackson signed the bill into law authorizing<|end-PI|> <|retrieval|> <|start-KFI|>the United States Exploring Expedition of 1838 – 42<|end-KFI|> . <|start-PI|>The expedition , led by Lt. Charles Wilkes , included<|end-PI|> <|retrieval|> <|start-KFI|>exploration and surveying of the Pacific Ocean and surrounding lands , including Antarctica<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context for the first mapping of Mount Jackson, and the KFI provides the specific date and team responsible. The <|retrieval|> token is inserted to indicate where additional information about the mapping expedition might be needed.",
"Ideal Knowledge": "A detailed timeline of Antarctic expeditions, focusing on the United States Antarctic Service (USAS) activities in 1940, particularly their topographic mapping efforts.",
"Key Factual Information": "November 1940 by a sledge party of the United States Antarctic Service ( USAS )",
"Prefix reformulated": "When was the first topographic mapping of Mount Jackson carried out?",
"Prefix-intent": "The first topographic mapping of Mount Jackson was carried out in",
"Verification": "The KFI directly provides the date and who conducted the first topographic mapping of Mount Jackson, as prompted by the PI."
},
{
"Explanation": "The PI introduces the topic of the mountain's estimated height, and the KFI provides the specific measurement. The <|retrieval|> token is inserted to indicate where additional information about mountain height estimates might be needed.",
"Ideal Knowledge": "A comprehensive list of Antarctic mountain heights, specifically including the initial estimates for Mount Jackson from the 1940 USAS expedition.",
"Key Factual Information": "4 @,@ 200 metres ( 13 @,@ 800 ft )",
"Prefix reformulated": "What was the estimated height of Mount Jackson?",
"Prefix-intent": "The height was estimated at",
"Verification": "The KFI directly provides the estimated height of Mount Jackson, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context for the mountain's initial naming, and the KFI provides the specific person and his position. The <|retrieval|> token is inserted to indicate where additional information about Ernest Gruening might be needed.",
"Ideal Knowledge": "Biographical information about Ernest Gruening, focusing on his role as Governor of the Alaska Territory and his connection to Antarctic exploration.",
"Key Factual Information": "the Governor of the Alaska Territory at that time , Ernest Gruening",
"Prefix reformulated": "Who was Mount Ernest Gruening named after?",
"Prefix-intent": "The mountain was named Mount Ernest Gruening after",
"Verification": "The KFI directly provides information about who Mount Ernest Gruening was named after, as prompted by the PI."
},
{
"Explanation": "The PI introduces the topic of the mountain's renaming, and the KFI provides the reason for the new name. The <|retrieval|> token is inserted to indicate where additional information about the renaming decision might be needed.",
"Ideal Knowledge": "Information about the naming conventions for Antarctic geographical features, particularly those named after U.S. Presidents, and specific details about why Andrew Jackson was chosen.",
"Key Factual Information": "the seventh President of the United States",
"Prefix reformulated": "Why did USAS rename Mount Ernest Gruening to Mount Jackson?",
"Prefix-intent": "USAS later renamed it Mount Jackson for",
"Verification": "The KFI directly provides the reason for renaming the mountain to Mount Jackson, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context for President Jackson's role in Antarctic exploration, and the KFI provides the specific expedition he authorized. The <|retrieval|> token is inserted to indicate where additional information about the expedition might be needed.",
"Ideal Knowledge": "Detailed information about the United States Exploring Expedition of 1838-42, including its authorization process, objectives, and significance in Antarctic exploration.",
"Key Factual Information": "the United States Exploring Expedition of 1838 – 42",
"Prefix reformulated": "What did President Jackson authorize through a bill he signed into law?",
"Prefix-intent": "President Jackson signed the bill into law authorizing",
"Verification": "The KFI directly provides information about what President Jackson authorized, as prompted by the PI."
},
{
"Explanation": "The PI introduces the expedition's leader and sets up a query about its activities, while the KFI provides the specific areas and tasks involved. The <|retrieval|> token is inserted to indicate where additional information about the expedition's scope might be needed.",
"Ideal Knowledge": "A comprehensive overview of the United States Exploring Expedition's activities, focusing on its Pacific Ocean and Antarctic explorations, surveying methods, and significant discoveries.",
"Key Factual Information": "exploration and surveying of the Pacific Ocean and surrounding lands , including Antarctica",
"Prefix reformulated": "What did the United States Exploring Expedition led by Lt. Charles Wilkes include?",
"Prefix-intent": "The expedition , led by Lt. Charles Wilkes , included",
"Verification": "The KFI directly provides information about what the expedition included, as prompted by the PI."
}
]
} |
In November 1947 , a Falkland Islands Dependencies Survey ( FIDS ) ground party based at Stonington Island observed Mount Jackson and estimated its height at 3 @,@ 050 metres ( 10 @,@ 010 ft ) , considerably lower and more accurate than the first estimate in 1940 .
| [
{
"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 November 1947 , a Falkland Islands Dependencies Survey ( FIDS ) ground party based at Stonington Island observed Mount Jackson and estimated its height at 3 @,@ 050 metres ( 10 @,@ 010 ft ) , considerably lower and more accurate than the first estimate in 1940 . \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 November 1947, a Falkland Islands Dependencies Survey (FIDS) ground party based at Stonington Island observed Mount Jackson and estimated its height at<|end-PI|> <|retrieval|> <|start-KFI|>3,050 metres (10,010 ft)<|end-KFI|>, <|start-PI|>considerably<|end-PI|> <|retrieval|> <|start-KFI|>lower and more accurate than the first estimate in 1940<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context for the height estimation of Mount Jackson, while the KFI provides the specific measurement. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement about the height estimation.",
"Ideal Knowledge": "A detailed record of the Falkland Islands Dependencies Survey (FIDS) from November 1947, including their observations and measurements of Mount Jackson, specifically mentioning the estimated height of 3,050 metres (10,010 ft).",
"Key Factual Information": "3,050 metres (10,010 ft)",
"Prefix reformulated": "What was the estimated height of Mount Jackson according to the Falkland Islands Dependencies Survey (FIDS) ground party in November 1947?",
"Prefix-intent": "In November 1947, a Falkland Islands Dependencies Survey (FIDS) ground party based at Stonington Island observed Mount Jackson and estimated its height at",
"Verification": "The KFI directly provides the height estimation requested in the PI."
},
{
"Explanation": "The PI provides context for the 1947 height estimation, while the KFI compares it to the previous estimate. The <|retrieval|> token is inserted to indicate where additional information about the comparison between estimates might be needed.",
"Ideal Knowledge": "A comparison of Mount Jackson height estimations, including the initial estimate from 1940 and the subsequent FIDS estimate from 1947, highlighting the differences in accuracy and measurements.",
"Key Factual Information": "lower and more accurate than the first estimate in 1940",
"Prefix reformulated": "How did the 1947 FIDS estimation of Mount Jackson's height compare to the previous estimate?",
"Prefix-intent": "In November 1947, a Falkland Islands Dependencies Survey (FIDS) ground party based at Stonington Island observed Mount Jackson and estimated its height at 3,050 metres (10,010 ft), considerably",
"Verification": "The KFI directly compares the 1947 estimation to the previous one, as prompted by the PI."
}
]
} |
After the Falkland Islands Dependencies Survey was renamed British Antarctic Survey in 1962 , a series of depots were developed , including one at Mount Jackson . BAS members were successful in making many first ascents of the mountains of the Antarctic Peninsula , and John Cunningham , who served at the BAS base at Adelaide Island , was the first to climb Mount Jackson . Traveling by dog sled for 640 kilometres ( 400 mi ) , his team reached the summit on 23 November 1964 .
| [
{
"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 the Falkland Islands Dependencies Survey was renamed British Antarctic Survey in 1962 , a series of depots were developed , including one at Mount Jackson . BAS members were successful in making many first ascents of the mountains of the Antarctic Peninsula , and John Cunningham , who served at the BAS base at Adelaide Island , was the first to climb Mount Jackson . Traveling by dog sled for 640 kilometres ( 400 mi ) , his team reached the summit on 23 November 1964 . \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 the Falkland Islands Dependencies Survey was renamed<|end-PI|> <|retrieval|> <|start-KFI|>British Antarctic Survey in 1962<|end-KFI|>, a series of depots were developed, including one at Mount Jackson. <|start-PI|>BAS members were successful in making many first ascents of the mountains of<|end-PI|> <|retrieval|> <|start-KFI|>the Antarctic Peninsula<|end-KFI|>, and <|start-PI|>John Cunningham, who served at the BAS base at Adelaide Island, was the first to climb<|end-PI|> <|retrieval|> <|start-KFI|>Mount Jackson<|end-KFI|>. <|start-PI|>Traveling by dog sled for<|end-PI|> <|retrieval|> <|start-KFI|>640 kilometres ( 400 mi )<|end-KFI|>, <|start-PI|>his team reached the summit on<|end-PI|> <|retrieval|> <|start-KFI|>23 November 1964<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context for the renaming, and the KFI provides the specific new name and date. The <|retrieval|> token is inserted to indicate where additional information about the renaming might be needed.",
"Ideal Knowledge": "A historical timeline of Antarctic research organizations, focusing on the renaming of the Falkland Islands Dependencies Survey to the British Antarctic Survey in 1962, including the reasons for the change and its implications.",
"Key Factual Information": "British Antarctic Survey in 1962",
"Prefix reformulated": "What was the Falkland Islands Dependencies Survey renamed to?",
"Prefix-intent": "After the Falkland Islands Dependencies Survey was renamed",
"Verification": "The KFI directly provides the new name and year of renaming as prompted by the PI."
},
{
"Explanation": "The PI introduces the accomplishments of BAS members, and the KFI provides the specific location of these achievements. The <|retrieval|> token is inserted to indicate where additional information about these first ascents might be needed.",
"Ideal Knowledge": "A comprehensive list of first ascents made by British Antarctic Survey members in the Antarctic Peninsula, including dates, names of climbers, and specific mountains conquered.",
"Key Factual Information": "the Antarctic Peninsula",
"Prefix reformulated": "Where did BAS members make many first ascents of mountains?",
"Prefix-intent": "BAS members were successful in making many first ascents of the mountains of",
"Verification": "The KFI directly specifies the location of the first ascents mentioned in the PI."
},
{
"Explanation": "The PI introduces John Cunningham and sets up the context for his achievement, while the KFI provides the specific mountain he climbed. The <|retrieval|> token is inserted to indicate where additional information about this climb might be needed.",
"Ideal Knowledge": "Detailed information about John Cunningham's expedition to Mount Jackson, including his role at the BAS base on Adelaide Island, the challenges of the climb, and the significance of this first ascent in Antarctic exploration history.",
"Key Factual Information": "Mount Jackson",
"Prefix reformulated": "Which mountain did John Cunningham, who served at the BAS base at Adelaide Island, first climb?",
"Prefix-intent": "John Cunningham, who served at the BAS base at Adelaide Island, was the first to climb",
"Verification": "The KFI directly provides the name of the mountain that John Cunningham first climbed, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of the travel method, and the KFI provides the specific distance covered. The <|retrieval|> token is inserted to indicate where additional information about the journey might be needed.",
"Ideal Knowledge": "Information on dog sledding expeditions in Antarctica, including typical distances covered, challenges faced, and the specific details of John Cunningham's journey to Mount Jackson, emphasizing the 640 km (400 mi) distance.",
"Key Factual Information": "640 kilometres ( 400 mi )",
"Prefix reformulated": "How far did John Cunningham's team travel by dog sled?",
"Prefix-intent": "Traveling by dog sled for",
"Verification": "The KFI directly provides the distance traveled by dog sled, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of reaching the summit, and the KFI provides the specific date. The <|retrieval|> token is inserted to indicate where additional information about this achievement might be needed.",
"Ideal Knowledge": "A detailed account of John Cunningham's expedition to Mount Jackson, focusing on the final ascent and the significance of reaching the summit on November 23, 1964, in the context of Antarctic exploration history.",
"Key Factual Information": "23 November 1964",
"Prefix reformulated": "When did John Cunningham's team reach the summit of Mount Jackson?",
"Prefix-intent": "his team reached the summit on",
"Verification": "The KFI directly provides the exact date of reaching the summit, as prompted by the PI."
}
]
} |
A precise height of the summit was determined by a BAS survey party who ascended the peak during the austral summer of 1996 – 97 .
| [
{
"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 precise height of the summit was determined by a BAS survey party who ascended the peak during the austral summer of 1996 – 97 . \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|>A precise height of the summit was determined by a BAS survey party who<|end-PI|> <|retrieval|> <|start-KFI|>ascended the peak during the austral summer of 1996 – 97<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces the fact that a precise height was determined by a BAS survey party, setting up the question of when this occurred. The KFI provides the specific timeframe. The <|retrieval|> token is inserted to indicate where additional information about the survey expedition might be needed.",
"Ideal Knowledge": "Detailed information about the British Antarctic Survey (BAS) expedition to measure the height of the summit, including the specific dates of the austral summer of 1996-97, the methods used for measurement, and the team members involved.",
"Key Factual Information": "ascended the peak during the austral summer of 1996 – 97",
"Prefix reformulated": "When did a BAS survey party determine the precise height of the summit?",
"Prefix-intent": "A precise height of the summit was determined by a BAS survey party who",
"Verification": "The KFI directly provides the time when the BAS survey party determined the height, as prompted by the reformulated PI."
}
]
} |
The mountain is 3 @,@ 184 metres ( 10 @,@ 446 ft ) in height , with a prominence of 2 @,@ 187 metres ( 7 @,@ 175 ft ) and a saddle DEM of 997 metres ( 3 @,@ 271 ft ) . Mount Jackson and the Welch Mountains demarcate the central Black Coast , which is dissected by many inlets and is bounded on the west by Dyer Plateau of central Palmer Land , with elevation ranging between 2 @,@ 000 – 3 @,@ 000 metres ( 6 @,@ 600 – 9 @,@ 800 ft ) , and on the west side of the central Black Coast . The two mountains rise above the ice shelf with reliefs of about 1 @,@ 200 – 1 @,@ 500 metres ( 3 @,@ 900 – 4 @,@ 900 ft ) towards the east . They are interconnected by icefall zones forming a plateau with steep snow slopes . The northern part of the mountain system , which extends over 1 @,@ 590 kilometres ( 990 mi ) , covers most of the Antarctic Peninsula , and Mount Jackson is its highest peak .
| [
{
"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 mountain is 3 @,@ 184 metres ( 10 @,@ 446 ft ) in height , with a prominence of 2 @,@ 187 metres ( 7 @,@ 175 ft ) and a saddle DEM of 997 metres ( 3 @,@ 271 ft ) . Mount Jackson and the Welch Mountains demarcate the central Black Coast , which is dissected by many inlets and is bounded on the west by Dyer Plateau of central Palmer Land , with elevation ranging between 2 @,@ 000 – 3 @,@ 000 metres ( 6 @,@ 600 – 9 @,@ 800 ft ) , and on the west side of the central Black Coast . The two mountains rise above the ice shelf with reliefs of about 1 @,@ 200 – 1 @,@ 500 metres ( 3 @,@ 900 – 4 @,@ 900 ft ) towards the east . They are interconnected by icefall zones forming a plateau with steep snow slopes . The northern part of the mountain system , which extends over 1 @,@ 590 kilometres ( 990 mi ) , covers most of the Antarctic Peninsula , and Mount Jackson is its highest peak . \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 mountain is 3 @,@ 184 metres ( 10 @,@ 446 ft ) in height , with a prominence of<|end-PI|> <|retrieval|> <|start-KFI|>2 @,@ 187 metres ( 7 @,@ 175 ft )<|end-KFI|> and a saddle DEM of 997 metres ( 3 @,@ 271 ft ) . <|start-PI|>Mount Jackson and the Welch Mountains demarcate the central Black Coast , which is<|end-PI|> <|retrieval|> <|start-KFI|>dissected by many inlets and is bounded on the west by Dyer Plateau of central Palmer Land<|end-KFI|> , with elevation ranging between 2 @,@ 000 – 3 @,@ 000 metres ( 6 @,@ 600 – 9 @,@ 800 ft ) , and on the west side of the central Black Coast . The two mountains rise above the ice shelf with reliefs of about 1 @,@ 200 – 1 @,@ 500 metres ( 3 @,@ 900 – 4 @,@ 900 ft ) towards the east . They are interconnected by icefall zones forming a plateau with steep snow slopes . <|start-PI|>The northern part of the mountain system , which extends over<|end-PI|> <|retrieval|> <|start-KFI|>1 @,@ 590 kilometres ( 990 mi ) , covers most of the Antarctic Peninsula , and Mount Jackson is its highest peak<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the mountain's height and sets up a query about its prominence, while the KFI provides the specific measurement. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "A detailed list of mountain measurements, including height, prominence, and saddle DEM, specifically for the mountain in question (likely Mount Jackson in Antarctica).",
"Key Factual Information": "2 @,@ 187 metres ( 7 @,@ 175 ft )",
"Prefix reformulated": "What is the prominence of the mountain that is 3,184 metres (10,446 ft) in height?",
"Prefix-intent": "The mountain is 3 @,@ 184 metres ( 10 @,@ 446 ft ) in height , with a prominence of",
"Verification": "The KFI directly provides the prominence information requested in the PI."
},
{
"Explanation": "The PI introduces Mount Jackson and the Welch Mountains in relation to the central Black Coast, while the KFI provides specific details about the coast's features. The <|retrieval|> token is inserted to indicate where additional geographical information might be needed.",
"Ideal Knowledge": "Geographical information about the central Black Coast in Antarctica, including its boundaries, topographical features, and surrounding landmarks.",
"Key Factual Information": "dissected by many inlets and is bounded on the west by Dyer Plateau of central Palmer Land",
"Prefix reformulated": "How is the central Black Coast, demarcated by Mount Jackson and the Welch Mountains, characterized?",
"Prefix-intent": "Mount Jackson and the Welch Mountains demarcate the central Black Coast , which is",
"Verification": "The KFI directly describes the characteristics of the central Black Coast, as prompted by the PI."
},
{
"Explanation": "The PI introduces the northern part of the mountain system and sets up a query about its extent, while the KFI provides specific measurements and additional relevant information. The <|retrieval|> token is inserted to indicate where additional geographical and topographical information might be needed.",
"Ideal Knowledge": "Detailed information about the Antarctic Peninsula's mountain system, including its extent, coverage area, and notable peaks, with specific focus on Mount Jackson.",
"Key Factual Information": "1 @,@ 590 kilometres ( 990 mi ) , covers most of the Antarctic Peninsula , and Mount Jackson is its highest peak",
"Prefix reformulated": "What is the extent of the northern part of the mountain system?",
"Prefix-intent": "The northern part of the mountain system , which extends over",
"Verification": "The KFI directly provides the extent of the mountain system and additional information about its coverage and highest peak, as prompted by the PI."
}
]
} |
Mount Jackson rises from its southeast flanks , displaying a steeple summit , while the north flank is occupied by a vast cirque . Supraglacial moraines on the mountain 's east side measure between 1 – 5 kilometres ( 0 @.@ 62 – 3 @.@ 11 mi ) in length and display boulders at their distal ends . Mount Jackson and the Rowley Massif are separated by Odom Inlet . Cline Glacier drains eastern Mount Jackson before arriving at Odom Inlet .
| [
{
"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 Mount Jackson rises from its southeast flanks , displaying a steeple summit , while the north flank is occupied by a vast cirque . Supraglacial moraines on the mountain 's east side measure between 1 – 5 kilometres ( 0 @.@ 62 – 3 @.@ 11 mi ) in length and display boulders at their distal ends . Mount Jackson and the Rowley Massif are separated by Odom Inlet . Cline Glacier drains eastern Mount Jackson before arriving at Odom Inlet . \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|>Mount Jackson rises from its southeast flanks , displaying<|end-PI|> <|retrieval|> <|start-KFI|>a steeple summit<|end-KFI|> , while <|start-PI|>the north flank is occupied by<|end-PI|> <|retrieval|> <|start-KFI|>a vast cirque<|end-KFI|> . <|start-PI|>Supraglacial moraines on the mountain 's east side measure between<|end-PI|> <|retrieval|> <|start-KFI|>1 – 5 kilometres ( 0 @.@ 62 – 3 @.@ 11 mi ) in length<|end-KFI|> and display boulders at their distal ends . <|start-PI|>Mount Jackson and the Rowley Massif are separated by<|end-PI|> <|retrieval|> <|start-KFI|>Odom Inlet<|end-KFI|> . <|start-PI|>Cline Glacier drains eastern Mount Jackson before arriving at<|end-PI|> <|retrieval|> <|start-KFI|>Odom Inlet<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces Mount Jackson and sets up a query about its appearance, while the KFI provides the specific description. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "Detailed description of Mount Jackson's topography, focusing on its summit shape and appearance from different angles.",
"Key Factual Information": "a steeple summit",
"Prefix reformulated": "What does Mount Jackson display on its southeast flanks?",
"Prefix-intent": "Mount Jackson rises from its southeast flanks , displaying",
"Verification": "The KFI directly describes what Mount Jackson displays, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the north flank of Mount Jackson, and the KFI provides the specific geological feature. The <|retrieval|> token is inserted to indicate where additional information about the cirque might be needed.",
"Ideal Knowledge": "Information on the geological features of Mount Jackson, specifically detailing the cirque on its north flank and its characteristics.",
"Key Factual Information": "a vast cirque",
"Prefix reformulated": "What occupies the north flank of Mount Jackson?",
"Prefix-intent": "the north flank is occupied by",
"Verification": "The KFI directly states what occupies the north flank, as prompted by the PI."
},
{
"Explanation": "The PI introduces the supraglacial moraines and sets up a query about their size, while the KFI provides the specific measurements. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "Detailed measurements of geological features on Mount Jackson, specifically focusing on the supraglacial moraines on its east side, including their dimensions in both metric and imperial units.",
"Key Factual Information": "1 – 5 kilometres ( 0 @.@ 62 – 3 @.@ 11 mi ) in length",
"Prefix reformulated": "What is the length of the supraglacial moraines on Mount Jackson's east side?",
"Prefix-intent": "Supraglacial moraines on the mountain 's east side measure between",
"Verification": "The KFI directly provides the length of the supraglacial moraines, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the geographical relationship between Mount Jackson and the Rowley Massif, and the KFI provides the specific feature that separates them. The <|retrieval|> token is inserted to indicate where additional information about Odom Inlet might be needed.",
"Ideal Knowledge": "Geographical information about the area surrounding Mount Jackson, including details about Odom Inlet and its position relative to Mount Jackson and the Rowley Massif.",
"Key Factual Information": "Odom Inlet",
"Prefix reformulated": "What separates Mount Jackson and the Rowley Massif?",
"Prefix-intent": "Mount Jackson and the Rowley Massif are separated by",
"Verification": "The KFI directly states what separates Mount Jackson and the Rowley Massif, as prompted by the PI."
},
{
"Explanation": "The PI introduces Cline Glacier and its path, and the KFI provides its destination. The <|retrieval|> token is inserted to indicate where additional information about the glacier's path might be needed.",
"Ideal Knowledge": "Information about the glacial system of Mount Jackson, specifically detailing the path of Cline Glacier from the eastern part of the mountain to Odom Inlet.",
"Key Factual Information": "Odom Inlet",
"Prefix reformulated": "Where does Cline Glacier arrive after draining eastern Mount Jackson?",
"Prefix-intent": "Cline Glacier drains eastern Mount Jackson before arriving at",
"Verification": "The KFI directly states where Cline Glacier arrives, as prompted by the PI."
}
]
} |
Mount Jackson 's geology was studied in 1972 as part of the Palmer Island investigations by a team of geologists . They identified various rock types and underlying volcanic activity . The relief of the west facing slopes is gentler when compared to the eastern side , characterized by a steep rocky face . Rock weathering is intense in the area spread with scree and felsenmeer . Fresh rock formations and bedrock disintegration are noted .
| [
{
"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 Mount Jackson 's geology was studied in 1972 as part of the Palmer Island investigations by a team of geologists . They identified various rock types and underlying volcanic activity . The relief of the west facing slopes is gentler when compared to the eastern side , characterized by a steep rocky face . Rock weathering is intense in the area spread with scree and felsenmeer . Fresh rock formations and bedrock disintegration are noted . \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
} |
Aretusa was a torpedo cruiser of the Partenope class built for the Italian Regia Marina ( Royal Navy ) in the 1880s . Laid down in June 1889 at the Cantiere navale fratelli Orlando shipyard , she was launched in March 1891 and was commissioned in September 1892 . Her main armament were her six torpedo tubes , which were supported by a battery of ten small @-@ caliber guns . Aretusa spent most of her career in the main Italian fleet , where she was primarily occupied with training exercises . At the start of the Italo @-@ Turkish War in September 1911 , she was assigned to the Red Sea Squadron in Italian Eritrea . She bombarded Ottoman positions in the Arabian Peninsula and took part in a blockade of the coast . Worn out by the end of the war in October 1912 , Aretusa was sold for scrap that December and broken up .
| [
{
"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 Aretusa was a torpedo cruiser of the Partenope class built for the Italian Regia Marina ( Royal Navy ) in the 1880s . Laid down in June 1889 at the Cantiere navale fratelli Orlando shipyard , she was launched in March 1891 and was commissioned in September 1892 . Her main armament were her six torpedo tubes , which were supported by a battery of ten small @-@ caliber guns . Aretusa spent most of her career in the main Italian fleet , where she was primarily occupied with training exercises . At the start of the Italo @-@ Turkish War in September 1911 , she was assigned to the Red Sea Squadron in Italian Eritrea . She bombarded Ottoman positions in the Arabian Peninsula and took part in a blockade of the coast . Worn out by the end of the war in October 1912 , Aretusa was sold for scrap that December and broken up . \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
} |
Aretusa was 73 @.@ 1 meters ( 239 ft 10 in ) long overall and had a beam of 8 @.@ 22 m ( 27 ft 0 in ) and an average draft of 3 @.@ 48 m ( 11 ft 5 in ) . She displaced 833 metric tons ( 820 long tons ; 918 short tons ) normally . Her propulsion system consisted of a pair of horizontal triple @-@ expansion steam engines , each driving a single screw propeller , with steam supplied by four coal @-@ fired locomotive boilers . Specific figures for Aretusa 's engine performance have not survived , but the ships of her class had top speeds of 18 @.@ 1 to 20 @.@ 8 knots ( 33 @.@ 5 to 38 @.@ 5 km / h ; 20 @.@ 8 to 23 @.@ 9 mph ) at 3 @,@ 884 to 4 @,@ 422 indicated horsepower ( 2 @,@ 896 to 3 @,@ 297 kW ) . The ship had a cruising radius of about 1 @,@ 800 nautical miles ( 3 @,@ 300 km ; 2 @,@ 100 mi ) at a speed of 10 knots ( 19 km / h ; 12 mph ) . She had a crew of between 96 and 121 .
| [
{
"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 Aretusa was 73 @.@ 1 meters ( 239 ft 10 in ) long overall and had a beam of 8 @.@ 22 m ( 27 ft 0 in ) and an average draft of 3 @.@ 48 m ( 11 ft 5 in ) . She displaced 833 metric tons ( 820 long tons ; 918 short tons ) normally . Her propulsion system consisted of a pair of horizontal triple @-@ expansion steam engines , each driving a single screw propeller , with steam supplied by four coal @-@ fired locomotive boilers . Specific figures for Aretusa 's engine performance have not survived , but the ships of her class had top speeds of 18 @.@ 1 to 20 @.@ 8 knots ( 33 @.@ 5 to 38 @.@ 5 km / h ; 20 @.@ 8 to 23 @.@ 9 mph ) at 3 @,@ 884 to 4 @,@ 422 indicated horsepower ( 2 @,@ 896 to 3 @,@ 297 kW ) . The ship had a cruising radius of about 1 @,@ 800 nautical miles ( 3 @,@ 300 km ; 2 @,@ 100 mi ) at a speed of 10 knots ( 19 km / h ; 12 mph ) . She had a crew of between 96 and 121 . \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
} |
Aretusa was armed with a main battery of one 120 mm ( 4 @.@ 7 in ) / 40 gun and six 57 mm ( 2 @.@ 2 in ) / 43 guns mounted singly.α She was also equipped with three 37 mm ( 1 @.@ 5 in ) / 20 guns in single mounts . Her primary offensive weapon was her five 450 mm ( 17 @.@ 7 in ) torpedo tubes . The ship was protected by an armored deck that was up to 1 @.@ 6 in ( 41 mm ) thick ; her conning tower was armored with the same thickness of steel plate .
| [
{
"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 Aretusa was armed with a main battery of one 120 mm ( 4 @.@ 7 in ) / 40 gun and six 57 mm ( 2 @.@ 2 in ) / 43 guns mounted singly.α She was also equipped with three 37 mm ( 1 @.@ 5 in ) / 20 guns in single mounts . Her primary offensive weapon was her five 450 mm ( 17 @.@ 7 in ) torpedo tubes . The ship was protected by an armored deck that was up to 1 @.@ 6 in ( 41 mm ) thick ; her conning tower was armored with the same thickness of steel plate . \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
} |
Aretusa was laid down at the Cantiere navale fratelli Orlando ( Orlando Brothers ' Shipyard ) in Livorno on 1 June 1889 , and was launched on 14 March 1891 . After fitting @-@ out work was completed , the ship was commissioned into the fleet on 1 September 1892 . During the 1893 fleet maneuvers , Aretusa served with the 3rd Division of the Reserve Squadron , along with the protected cruisers Vesuvio and Ettore Fieramosca and four torpedo boats . During the maneuvers , which lasted from 6 August to 5 September , the ships of the Reserve Squadron defended against a simulated attack by the Active Squadron , which gamed a French attack on the Italian fleet . In 1895 , Aretusa was stationed in the 2nd Maritime Department , split between Taranto and Naples , along with most of the torpedo cruisers in the Italian fleet . These included her sister ships Partenope , Minerva , Euridice , Iride , Urania , and Caprera , the four Goito @-@ class cruisers , and Tripoli . As of 1898 , Aretusa was assigned to the Active Squadron , with included the ironclads Sicilia and Sardegna and two other cruisers .
| [
{
"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 Aretusa was laid down at the Cantiere navale fratelli Orlando ( Orlando Brothers ' Shipyard ) in Livorno on 1 June 1889 , and was launched on 14 March 1891 . After fitting @-@ out work was completed , the ship was commissioned into the fleet on 1 September 1892 . During the 1893 fleet maneuvers , Aretusa served with the 3rd Division of the Reserve Squadron , along with the protected cruisers Vesuvio and Ettore Fieramosca and four torpedo boats . During the maneuvers , which lasted from 6 August to 5 September , the ships of the Reserve Squadron defended against a simulated attack by the Active Squadron , which gamed a French attack on the Italian fleet . In 1895 , Aretusa was stationed in the 2nd Maritime Department , split between Taranto and Naples , along with most of the torpedo cruisers in the Italian fleet . These included her sister ships Partenope , Minerva , Euridice , Iride , Urania , and Caprera , the four Goito @-@ class cruisers , and Tripoli . As of 1898 , Aretusa was assigned to the Active Squadron , with included the ironclads Sicilia and Sardegna and two other cruisers . \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|>Aretusa was laid down at the Cantiere navale fratelli Orlando ( Orlando Brothers ' Shipyard ) in Livorno on<|end-PI|> <|retrieval|> <|start-KFI|>1 June 1889<|end-KFI|> , and <|start-PI|>Aretusa was launched on<|end-PI|> <|retrieval|> <|start-KFI|>14 March 1891<|end-KFI|> . <|start-PI|>After fitting @-@ out work was completed , the ship was commissioned into the fleet on<|end-PI|> <|retrieval|> <|start-KFI|>1 September 1892<|end-KFI|> . <|start-PI|>During the 1893 fleet maneuvers , Aretusa served with the 3rd Division of the Reserve Squadron , along with<|end-PI|> <|retrieval|> <|start-KFI|>the protected cruisers Vesuvio and Ettore Fieramosca and four torpedo boats<|end-KFI|> . During the maneuvers , which lasted from 6 August to 5 September , the ships of the Reserve Squadron defended against a simulated attack by the Active Squadron , which gamed a French attack on the Italian fleet . <|start-PI|>In 1895 , Aretusa was stationed in the 2nd Maritime Department , split between Taranto and Naples , along with<|end-PI|> <|retrieval|> <|start-KFI|>most of the torpedo cruisers in the Italian fleet . These included her sister ships Partenope , Minerva , Euridice , Iride , Urania , and Caprera , the four Goito @-@ class cruisers , and Tripoli<|end-KFI|> . <|start-PI|>As of 1898 , Aretusa was assigned to the Active Squadron , with included<|end-PI|> <|retrieval|> <|start-KFI|>the ironclads Sicilia and Sardegna and two other cruisers<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces Aretusa and sets up a query about its construction 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 timeline of Italian naval ship construction in the late 19th century, specifically mentioning the laying down of Aretusa at the Orlando Brothers' Shipyard in Livorno on June 1, 1889.",
"Key Factual Information": "1 June 1889",
"Prefix reformulated": "When was Aretusa laid down at the Cantiere navale fratelli Orlando in Livorno?",
"Prefix-intent": "Aretusa was laid down at the Cantiere navale fratelli Orlando ( Orlando Brothers ' Shipyard ) in Livorno on",
"Verification": "The KFI directly provides the date requested in the PI."
},
{
"Explanation": "The PI sets up a query about Aretusa's launch date, and the KFI provides the specific date. The <|retrieval|> token is inserted to indicate where additional information about the ship's construction timeline might be needed.",
"Ideal Knowledge": "A chronological list of Italian naval ship launches in the late 19th century, highlighting the launch of Aretusa on March 14, 1891.",
"Key Factual Information": "14 March 1891",
"Prefix reformulated": "On what date was Aretusa launched?",
"Prefix-intent": "Aretusa was launched on",
"Verification": "The KFI directly provides the launch date requested in the PI."
},
{
"Explanation": "The PI introduces the completion of fitting-out work and sets up a query about the commissioning 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 Italian naval ship commissionings in the early 1890s, specifically mentioning Aretusa's entry into service on September 1, 1892, following its fitting-out period.",
"Key Factual Information": "1 September 1892",
"Prefix reformulated": "When was Aretusa commissioned into the fleet after fitting-out work was completed?",
"Prefix-intent": "After fitting @-@ out work was completed , the ship was commissioned into the fleet on",
"Verification": "The KFI directly provides the commissioning date requested in the PI."
},
{
"Explanation": "The PI sets up information about Aretusa's service during the 1893 fleet maneuvers and prompts for information about other ships in its division, which the KFI provides. The <|retrieval|> token is inserted to indicate where additional information about the fleet composition might be needed.",
"Ideal Knowledge": "A detailed composition of the Italian Reserve Squadron during the 1893 fleet maneuvers, specifically listing the ships in the 3rd Division including Aretusa, Vesuvio, Ettore Fieramosca, and four torpedo boats.",
"Key Factual Information": "the protected cruisers Vesuvio and Ettore Fieramosca and four torpedo boats",
"Prefix reformulated": "Which ships served with Aretusa in the 3rd Division of the Reserve Squadron during the 1893 fleet maneuvers?",
"Prefix-intent": "During the 1893 fleet maneuvers , Aretusa served with the 3rd Division of the Reserve Squadron , along with",
"Verification": "The KFI directly lists the ships that served alongside Aretusa, as prompted by the PI."
},
{
"Explanation": "The PI introduces Aretusa's station in 1895 and prompts for information about other ships stationed with it, which the KFI provides in detail. The <|retrieval|> token is inserted to indicate where additional information about the fleet composition might be needed.",
"Ideal Knowledge": "A comprehensive list of Italian naval ships stationed in the 2nd Maritime Department in 1895, detailing the types of ships and their specific names, with a focus on torpedo cruisers and their assignments between Taranto and Naples.",
"Key Factual Information": "most of the torpedo cruisers in the Italian fleet . These included her sister ships Partenope , Minerva , Euridice , Iride , Urania , and Caprera , the four Goito @-@ class cruisers , and Tripoli",
"Prefix reformulated": "Which ships were stationed with Aretusa in the 2nd Maritime Department in 1895?",
"Prefix-intent": "In 1895 , Aretusa was stationed in the 2nd Maritime Department , split between Taranto and Naples , along with",
"Verification": "The KFI directly lists the ships stationed with Aretusa, as prompted by the PI."
},
{
"Explanation": "The PI introduces Aretusa's assignment in 1898 and prompts for information about other ships in the Active Squadron, which the KFI provides. The <|retrieval|> token is inserted to indicate where additional information about the squadron composition might be needed.",
"Ideal Knowledge": "A detailed composition of the Italian Active Squadron in 1898, listing all ships assigned to it, with specific mention of Aretusa, Sicilia, Sardegna, and the types of other vessels included.",
"Key Factual Information": "the ironclads Sicilia and Sardegna and two other cruisers",
"Prefix reformulated": "Which ships were included in the Active Squadron with Aretusa in 1898?",
"Prefix-intent": "As of 1898 , Aretusa was assigned to the Active Squadron , with included",
"Verification": "The KFI directly lists the ships assigned to the Active Squadron with Aretusa, as prompted by the PI."
}
]
} |
At the start of the Italo @-@ Turkish War in September 1911 , Aretusa was stationed in Italian Eritrea in the Red Sea Squadron . Italian naval forces in the region also included five protected cruisers and several smaller vessels . Shortly after the start of the war on 2 October , Aretusa and the gunboat Volturno encountered the Ottoman torpedo cruiser Peyk @-@ i Şevket off Al Hudaydah . In a short engagement , the Italians vessels forced the Ottoman ship to flee into Al Hudaydah , bombarded the port facilities , and then withdrew . Peyk @-@ i Şevket was later interned in British @-@ controlled Suez The threat of an Ottoman attack from the Arabian Peninsula led the Italian High Command to reinforce the Red Sea Squadron ; the additional ships included another cruiser and several destroyers . The protected cruiser Piemonte and two destroyers annihilated a force of seven Ottoman gunboats in the Battle of Kunfuda Bay on 7 January 1912 .
| [
{
"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 start of the Italo @-@ Turkish War in September 1911 , Aretusa was stationed in Italian Eritrea in the Red Sea Squadron . Italian naval forces in the region also included five protected cruisers and several smaller vessels . Shortly after the start of the war on 2 October , Aretusa and the gunboat Volturno encountered the Ottoman torpedo cruiser Peyk @-@ i Şevket off Al Hudaydah . In a short engagement , the Italians vessels forced the Ottoman ship to flee into Al Hudaydah , bombarded the port facilities , and then withdrew . Peyk @-@ i Şevket was later interned in British @-@ controlled Suez The threat of an Ottoman attack from the Arabian Peninsula led the Italian High Command to reinforce the Red Sea Squadron ; the additional ships included another cruiser and several destroyers . The protected cruiser Piemonte and two destroyers annihilated a force of seven Ottoman gunboats in the Battle of Kunfuda Bay on 7 January 1912 . \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 start of the Italo @-@ Turkish War in September 1911 , <|start-PI|> Aretusa was stationed in <|end-PI|> <|retrieval|> <|start-KFI|> Italian Eritrea in the Red Sea Squadron <|end-KFI|> . <|start-PI|> Italian naval forces in the region also included <|end-PI|> <|retrieval|> <|start-KFI|> five protected cruisers and several smaller vessels <|end-KFI|> . <|start-PI|> Shortly after the start of the war on 2 October , Aretusa and the gunboat Volturno encountered <|end-PI|> <|retrieval|> <|start-KFI|> the Ottoman torpedo cruiser Peyk @-@ i Şevket off Al Hudaydah <|end-KFI|> . In a short engagement , the Italians vessels forced the Ottoman ship to flee into Al Hudaydah , bombarded the port facilities , and then withdrew . Peyk @-@ i Şevket was later interned in British @-@ controlled Suez <|start-PI|> The threat of an Ottoman attack from the Arabian Peninsula led the Italian High Command to <|end-PI|> <|retrieval|> <|start-KFI|> reinforce the Red Sea Squadron ; the additional ships included another cruiser and several destroyers <|end-KFI|> . The protected cruiser Piemonte and two destroyers annihilated a force of seven Ottoman gunboats in the Battle of Kunfuda Bay on 7 January 1912 .",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of the war and introduces Aretusa, while the KFI provides its specific location. The <|retrieval|> token is inserted to indicate where additional information about naval deployments might be needed.",
"Ideal Knowledge": "A detailed account of Italian naval deployments at the start of the Italo-Turkish War, specifically mentioning the location and composition of the Red Sea Squadron.",
"Key Factual Information": "Italian Eritrea in the Red Sea Squadron",
"Prefix reformulated": "Where was Aretusa stationed at the start of the Italo-Turkish War in September 1911?",
"Prefix-intent": "At the start of the Italo @-@ Turkish War in September 1911 , Aretusa was stationed in",
"Verification": "The KFI directly provides the location information requested in the PI."
},
{
"Explanation": "The PI introduces the topic of additional Italian naval forces, and the KFI provides specific information about their composition. The <|retrieval|> token is inserted to indicate where more detailed information about these forces might be needed.",
"Ideal Knowledge": "A comprehensive list of Italian naval assets deployed in the Red Sea region during the Italo-Turkish War, including details on protected cruisers and smaller vessels.",
"Key Factual Information": "five protected cruisers and several smaller vessels",
"Prefix reformulated": "What other Italian naval forces were in the region during the Italo-Turkish War?",
"Prefix-intent": "Italian naval forces in the region also included",
"Verification": "The KFI directly lists the additional naval forces mentioned in the PI."
},
{
"Explanation": "The PI sets up the context of a naval encounter, and the KFI provides specific information about the Ottoman ship involved. The <|retrieval|> token is inserted to indicate where additional information about this encounter might be needed.",
"Ideal Knowledge": "Detailed information about naval engagements in the Red Sea during the early days of the Italo-Turkish War, specifically mentioning the encounter between Italian and Ottoman ships near Al Hudaydah.",
"Key Factual Information": "the Ottoman torpedo cruiser Peyk @-@ i Şevket off Al Hudaydah",
"Prefix reformulated": "Which Ottoman ship did Aretusa and the gunboat Volturno encounter shortly after the start of the war on 2 October?",
"Prefix-intent": "Shortly after the start of the war on 2 October , Aretusa and the gunboat Volturno encountered",
"Verification": "The KFI directly identifies the Ottoman ship encountered, as prompted by the PI."
},
{
"Explanation": "The PI introduces the threat and sets up a query about the Italian response, while the KFI provides specific information about the reinforcements. The <|retrieval|> token is inserted to indicate where additional strategic information might be needed.",
"Ideal Knowledge": "Information on Italian naval strategy in the Red Sea during the Italo-Turkish War, including details on reinforcements sent to counter potential Ottoman attacks from the Arabian Peninsula.",
"Key Factual Information": "reinforce the Red Sea Squadron ; the additional ships included another cruiser and several destroyers",
"Prefix reformulated": "How did the Italian High Command respond to the threat of an Ottoman attack from the Arabian Peninsula?",
"Prefix-intent": "The threat of an Ottoman attack from the Arabian Peninsula led the Italian High Command to",
"Verification": "The KFI directly describes the Italian response to the threat mentioned in the PI."
}
]
} |
Following the neutralization of Ottoman naval forces in the region , Aretusa and the rest of the Italian ships then commenced a bombardment campaign against the Turkish ports in the Red Sea before declaring a blockade of the city of Al Hudaydah on 26 January . On 27 July and 12 August , Aretusa , her sister ship Caprera , and Piemonte conducted two bombardments of Al Hudaydah . During the second attack , they destroyed an Ottoman ammunition dump . With the threat of an Ottoman attack greatly reduced , the High Command thereafter began to withdraw forces from the Red Sea Squadron . By the end of August , the unit was reduced to three protected cruisers , Aretusa , Caprera and two auxiliaries . On 14 October , the Ottoman government agreed to sign a peace treaty , ending the war . Aretusa 's career ended shortly thereafter ; the Regia Marina discarded the ship in December and she was subsequently broken up for scrap .
| [
{
"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 Following the neutralization of Ottoman naval forces in the region , Aretusa and the rest of the Italian ships then commenced a bombardment campaign against the Turkish ports in the Red Sea before declaring a blockade of the city of Al Hudaydah on 26 January . On 27 July and 12 August , Aretusa , her sister ship Caprera , and Piemonte conducted two bombardments of Al Hudaydah . During the second attack , they destroyed an Ottoman ammunition dump . With the threat of an Ottoman attack greatly reduced , the High Command thereafter began to withdraw forces from the Red Sea Squadron . By the end of August , the unit was reduced to three protected cruisers , Aretusa , Caprera and two auxiliaries . On 14 October , the Ottoman government agreed to sign a peace treaty , ending the war . Aretusa 's career ended shortly thereafter ; the Regia Marina discarded the ship in December and she was subsequently broken up for scrap . \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|>Following the neutralization of Ottoman naval forces in the region, Aretusa and the rest of the Italian ships then commenced<|end-PI|> <|retrieval|> <|start-KFI|>a bombardment campaign against the Turkish ports in the Red Sea before declaring a blockade of the city of Al Hudaydah on 26 January<|end-KFI|>. <|start-PI|>On 27 July and 12 August, Aretusa, her sister ship Caprera, and Piemonte conducted<|end-PI|> <|retrieval|> <|start-KFI|>two bombardments of Al Hudaydah. During the second attack, they destroyed an Ottoman ammunition dump<|end-KFI|>. <|start-PI|>With the threat of an Ottoman attack greatly reduced, the High Command thereafter began<|end-PI|> <|retrieval|> <|start-KFI|>to withdraw forces from the Red Sea Squadron. By the end of August, the unit was reduced to three protected cruisers, Aretusa, Caprera and two auxiliaries<|end-KFI|>. <|start-PI|>On 14 October, the Ottoman government agreed<|end-PI|> <|retrieval|> <|start-KFI|>to sign a peace treaty, ending the war<|end-KFI|>. <|start-PI|>Aretusa's career ended shortly thereafter; the Regia Marina<|end-PI|> <|retrieval|> <|start-KFI|>discarded the ship in December and she was subsequently broken up for scrap<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of Italian naval actions after neutralizing Ottoman forces, while the KFI provides specific details about the bombardment campaign and blockade. The <|retrieval|> token is inserted to indicate where additional information about these naval operations might be needed.",
"Ideal Knowledge": "Detailed information about Italian naval operations in the Red Sea during the Italo-Turkish War, including specific actions taken against Ottoman forces and ports.",
"Key Factual Information": "a bombardment campaign against the Turkish ports in the Red Sea before declaring a blockade of the city of Al Hudaydah on 26 January",
"Prefix reformulated": "What did Aretusa and the Italian ships do after neutralizing Ottoman naval forces?",
"Prefix-intent": "Following the neutralization of Ottoman naval forces in the region, Aretusa and the rest of the Italian ships then commenced",
"Verification": "The KFI directly describes the actions taken by Aretusa and Italian ships, as prompted by the PI."
},
{
"Explanation": "The PI introduces the ships involved and the dates of their actions, while the KFI provides details about the bombardments and their results. The <|retrieval|> token is inserted to indicate where additional information about these specific naval engagements might be needed.",
"Ideal Knowledge": "Detailed accounts of the specific bombardments carried out by Italian ships against Al Hudaydah, including dates, participating ships, and the outcomes of these attacks.",
"Key Factual Information": "two bombardments of Al Hudaydah. During the second attack, they destroyed an Ottoman ammunition dump",
"Prefix reformulated": "What actions did Aretusa, Caprera, and Piemonte take on 27 July and 12 August?",
"Prefix-intent": "On 27 July and 12 August, Aretusa, her sister ship Caprera, and Piemonte conducted",
"Verification": "The KFI directly describes the actions taken by the ships on the specified dates, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of reduced Ottoman threat, while the KFI provides specific details about the withdrawal of forces and the remaining composition of the Red Sea Squadron. The <|retrieval|> token is inserted to indicate where additional information about these strategic decisions might be needed.",
"Ideal Knowledge": "Information on the Italian naval strategy in the Red Sea during the later stages of the Italo-Turkish War, including details on force reductions and the composition of remaining naval units.",
"Key Factual Information": "to withdraw forces from the Red Sea Squadron. By the end of August, the unit was reduced to three protected cruisers, Aretusa, Caprera and two auxiliaries",
"Prefix reformulated": "What did the Italian High Command do after the Ottoman threat was reduced?",
"Prefix-intent": "With the threat of an Ottoman attack greatly reduced, the High Command thereafter began",
"Verification": "The KFI directly describes the actions taken by the Italian High Command and the resulting force reduction, as prompted by the PI."
},
{
"Explanation": "The PI introduces the date and the Ottoman government's action, while the KFI provides the specific agreement made and its significance. The <|retrieval|> token is inserted to indicate where additional information about the peace treaty might be needed.",
"Ideal Knowledge": "Details about the peace negotiations and treaty that ended the Italo-Turkish War, including the date of agreement and key terms.",
"Key Factual Information": "to sign a peace treaty, ending the war",
"Prefix reformulated": "What did the Ottoman government agree to on 14 October?",
"Prefix-intent": "On 14 October, the Ottoman government agreed",
"Verification": "The KFI directly states the agreement made by the Ottoman government, as prompted by the PI."
},
{
"Explanation": "The PI introduces the end of Aretusa's career and the Regia Marina's involvement, while the KFI provides specific details about the ship's fate. The <|retrieval|> token is inserted to indicate where additional information about the ship's decommissioning might be needed.",
"Ideal Knowledge": "Information on the post-war fate of Italian naval ships, specifically the decommissioning and scrapping process for Aretusa.",
"Key Factual Information": "discarded the ship in December and she was subsequently broken up for scrap",
"Prefix reformulated": "What did the Regia Marina do with Aretusa after the war?",
"Prefix-intent": "Aretusa's career ended shortly thereafter; the Regia Marina",
"Verification": "The KFI directly describes the fate of Aretusa after the war, as prompted by the PI."
}
]
} |
M @-@ 6 , or the Paul B. Henry Freeway , is a 19 @.@ 696 @-@ mile ( 31 @.@ 698 km ) freeway and state trunkline highway in the United States that serves portions of southern Kent and eastern Ottawa counties south of Grand Rapids , Michigan . Although the freeway is named for Paul B. Henry , local residents and the press continue to use the original name , South Beltline as well on occasion . The freeway connects Interstate 196 ( I @-@ 196 ) on the west with I @-@ 96 on the east . M @-@ 6 also provides a connection to U.S. Highway 131 ( US 131 ) in the middle of its corridor while running through several townships on the south side of the Grand Rapids metropolitan area in Western Michigan . Each end is in a rural area while the central section has suburban development along the trunkline .
| [
{
"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 M @-@ 6 , or the Paul B. Henry Freeway , is a 19 @.@ 696 @-@ mile ( 31 @.@ 698 km ) freeway and state trunkline highway in the United States that serves portions of southern Kent and eastern Ottawa counties south of Grand Rapids , Michigan . Although the freeway is named for Paul B. Henry , local residents and the press continue to use the original name , South Beltline as well on occasion . The freeway connects Interstate 196 ( I @-@ 196 ) on the west with I @-@ 96 on the east . M @-@ 6 also provides a connection to U.S. Highway 131 ( US 131 ) in the middle of its corridor while running through several townships on the south side of the Grand Rapids metropolitan area in Western Michigan . Each end is in a rural area while the central section has suburban development along the trunkline . \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|>M @-@ 6 , or the Paul B. Henry Freeway , is a<|end-PI|> <|retrieval|> <|start-KFI|>19 @.@ 696 @-@ mile ( 31 @.@ 698 km ) freeway and state trunkline highway<|end-KFI|> in the United States that serves portions of southern Kent and eastern Ottawa counties south of Grand Rapids , Michigan . <|start-PI|>Although the freeway is named for Paul B. Henry , local residents and the press continue to use<|end-PI|> <|retrieval|> <|start-KFI|>the original name , South Beltline as well on occasion<|end-KFI|> . <|start-PI|>The freeway connects<|end-PI|> <|retrieval|> <|start-KFI|>Interstate 196 ( I @-@ 196 ) on the west with I @-@ 96 on the east<|end-KFI|> . <|start-PI|>M @-@ 6 also provides a connection to<|end-PI|> <|retrieval|> <|start-KFI|>U.S. Highway 131 ( US 131 ) in the middle of its corridor<|end-KFI|> while running through several townships on the south side of the Grand Rapids metropolitan area in Western Michigan . Each end is in a rural area while the central section has suburban development along the trunkline .",
"none": null,
"results": [
{
"Explanation": "The PI introduces M-6 and sets up a query about its length, while the KFI provides the specific measurement. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "A detailed description of M-6 (Paul B. Henry Freeway) including its exact length in miles and kilometers, its classification as a freeway and state trunkline highway, and its location in Michigan.",
"Key Factual Information": "19 @.@ 696 @-@ mile ( 31 @.@ 698 km ) freeway and state trunkline highway",
"Prefix reformulated": "What is the length of M-6, also known as the Paul B. Henry Freeway?",
"Prefix-intent": "M @-@ 6 , or the Paul B. Henry Freeway , is a",
"Verification": "The KFI directly provides the length information requested in the PI."
},
{
"Explanation": "The PI sets up a query about an alternative name for the freeway, and the KFI provides the specific name used locally. The <|retrieval|> token is inserted to indicate where additional information about local naming conventions might be needed.",
"Ideal Knowledge": "Information about the naming history of M-6, including its official name as Paul B. Henry Freeway and the continued use of its original name, South Beltline, by local residents and media.",
"Key Factual Information": "the original name , South Beltline as well on occasion",
"Prefix reformulated": "What alternative name do local residents and the press use for the Paul B. Henry Freeway?",
"Prefix-intent": "Although the freeway is named for Paul B. Henry , local residents and the press continue to use",
"Verification": "The KFI directly provides the alternative name used by locals, as prompted by the PI."
},
{
"Explanation": "The PI introduces the concept of the freeway's connections, and the KFI specifies which highways it connects. The <|retrieval|> token is inserted to indicate where additional information about the highway system might be needed.",
"Ideal Knowledge": "A detailed map or description of the highway system in the Grand Rapids area, showing how M-6 connects I-196 on the west to I-96 on the east, including their relative positions and intersections.",
"Key Factual Information": "Interstate 196 ( I @-@ 196 ) on the west with I @-@ 96 on the east",
"Prefix reformulated": "What highways does M-6 (Paul B. Henry Freeway) connect?",
"Prefix-intent": "The freeway connects",
"Verification": "The KFI directly provides the highway connections requested in the PI."
},
{
"Explanation": "The PI sets up a query about an additional connection, and the KFI provides the specific highway and its location along M-6. The <|retrieval|> token is inserted to indicate where additional information about this intersection might be needed.",
"Ideal Knowledge": "Detailed information about the intersection of M-6 and US 131, including its location along the M-6 corridor and its significance in the local highway network.",
"Key Factual Information": "U.S. Highway 131 ( US 131 ) in the middle of its corridor",
"Prefix reformulated": "What additional highway does M-6 connect to?",
"Prefix-intent": "M @-@ 6 also provides a connection to",
"Verification": "The KFI directly provides the additional highway connection requested in the PI."
}
]
} |
The freeway was originally conceived in the 1960s . It took 32 years to approve , plan , finance , and build the freeway from the time that the state first authorized funding in 1972 to the time of the ribbon @-@ cutting ceremony in 2004 that opened the South Beltline to traffic . The project cost around $ 700 million or around $ 35 million per mile ( approximately $ 22 million per kilometer ) . Initial construction started in November 1997 , with the first phase opened in November 2001 . The full freeway was opened in November 2004 . The first phase of construction was completed in asphalt , while the second and third phases were built in concrete . The project was built with two firsts : the first single @-@ point urban interchange ( SPUI ; / ˈspuːiː / ) in Michigan , and a new technique to apply the pavement markings , embedding them into the concrete to reduce the chance of a snowplow scraping them off . In advance of the opening of the freeway to traffic , the Michigan Department of Transportation ( MDOT ) allowed the public to walk or bike the South Beltline in an open @-@ house event called the " Southbelt Shuffle " .
| [
{
"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 freeway was originally conceived in the 1960s . It took 32 years to approve , plan , finance , and build the freeway from the time that the state first authorized funding in 1972 to the time of the ribbon @-@ cutting ceremony in 2004 that opened the South Beltline to traffic . The project cost around $ 700 million or around $ 35 million per mile ( approximately $ 22 million per kilometer ) . Initial construction started in November 1997 , with the first phase opened in November 2001 . The full freeway was opened in November 2004 . The first phase of construction was completed in asphalt , while the second and third phases were built in concrete . The project was built with two firsts : the first single @-@ point urban interchange ( SPUI ; / ˈspuːiː / ) in Michigan , and a new technique to apply the pavement markings , embedding them into the concrete to reduce the chance of a snowplow scraping them off . In advance of the opening of the freeway to traffic , the Michigan Department of Transportation ( MDOT ) allowed the public to walk or bike the South Beltline in an open @-@ house event called the \" Southbelt Shuffle \" . \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 freeway was originally conceived in<|end-PI|> <|retrieval|> <|start-KFI|>the 1960s<|end-KFI|>. <|start-PI|>It took 32 years to approve , plan , finance , and build the freeway from the time that the state first authorized funding in<|end-PI|> <|retrieval|> <|start-KFI|>1972 to the time of the ribbon @-@ cutting ceremony in 2004 that opened the South Beltline to traffic<|end-KFI|>. <|start-PI|>The project cost around<|end-PI|> <|retrieval|> <|start-KFI|>$ 700 million or around $ 35 million per mile ( approximately $ 22 million per kilometer )<|end-KFI|>. <|start-PI|>Initial construction started in<|end-PI|> <|retrieval|> <|start-KFI|>November 1997 , with the first phase opened in November 2001<|end-KFI|>. <|start-PI|>The full freeway was opened in<|end-PI|> <|retrieval|> <|start-KFI|>November 2004<|end-KFI|>. The first phase of construction was completed in asphalt , while the second and third phases were built in concrete. <|start-PI|>The project was built with two firsts :<|end-PI|> <|retrieval|> <|start-KFI|>the first single @-@ point urban interchange ( SPUI ; / ˈspuːiː / ) in Michigan , and a new technique to apply the pavement markings , embedding them into the concrete to reduce the chance of a snowplow scraping them off<|end-KFI|>. <|start-PI|>In advance of the opening of the freeway to traffic , the Michigan Department of Transportation ( MDOT ) allowed the public to<|end-PI|> <|retrieval|> <|start-KFI|>walk or bike the South Beltline in an open @-@ house event called the \" Southbelt Shuffle \"<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces the freeway's conception, and the KFI provides the specific time period. The <|retrieval|> token is inserted to indicate where additional historical context might be needed.",
"Ideal Knowledge": "A timeline of the freeway's conception and development, specifically mentioning its origin in the 1960s and subsequent milestones.",
"Key Factual Information": "the 1960s",
"Prefix reformulated": "When was the freeway originally conceived?",
"Prefix-intent": "The freeway was originally conceived in",
"Verification": "The KFI directly provides the time period requested in the PI."
},
{
"Explanation": "The PI sets up the timeframe for the freeway project, and the KFI provides the specific start and end dates. The <|retrieval|> token is inserted to indicate where additional information about the project timeline might be needed.",
"Ideal Knowledge": "Detailed timeline of the freeway project, including key dates such as funding authorization in 1972 and the opening ceremony in 2004, along with major milestones in between.",
"Key Factual Information": "1972 to the time of the ribbon @-@ cutting ceremony in 2004 that opened the South Beltline to traffic",
"Prefix reformulated": "When did the state first authorize funding for the freeway?",
"Prefix-intent": "It took 32 years to approve , plan , finance , and build the freeway from the time that the state first authorized funding in",
"Verification": "The KFI directly provides the start and end dates of the freeway project as prompted by the PI."
},
{
"Explanation": "The PI introduces the project cost, and the KFI provides specific figures. The <|retrieval|> token is inserted to indicate where additional financial information might be needed.",
"Ideal Knowledge": "Detailed breakdown of the freeway project costs, including total cost and cost per unit of distance, with comparisons to similar projects for context.",
"Key Factual Information": "$ 700 million or around $ 35 million per mile ( approximately $ 22 million per kilometer )",
"Prefix reformulated": "How much did the freeway project cost?",
"Prefix-intent": "The project cost around",
"Verification": "The KFI directly provides the cost information requested in the PI."
},
{
"Explanation": "The PI asks about the start of construction, and the KFI provides the specific date and additional information about the first phase. The <|retrieval|> token is inserted to indicate where additional information about the construction timeline might be needed.",
"Ideal Knowledge": "A detailed construction timeline of the freeway project, highlighting key dates such as the start of construction in November 1997 and the opening of the first phase in November 2001.",
"Key Factual Information": "November 1997 , with the first phase opened in November 2001",
"Prefix reformulated": "When did the initial construction of the freeway start?",
"Prefix-intent": "Initial construction started in",
"Verification": "The KFI directly provides the start date of construction and the completion date of the first phase, as prompted by the PI."
},
{
"Explanation": "The PI asks about the opening of the full freeway, and the KFI provides the specific date. The <|retrieval|> token is inserted to indicate where additional information about the freeway's opening might be needed.",
"Ideal Knowledge": "Information about the freeway's completion and opening, including the specific date in November 2004 and any associated events or ceremonies.",
"Key Factual Information": "November 2004",
"Prefix reformulated": "When was the full freeway opened?",
"Prefix-intent": "The full freeway was opened in",
"Verification": "The KFI directly provides the opening date of the full freeway as requested in the PI."
},
{
"Explanation": "The PI introduces the unique aspects of the project, and the KFI provides specific details about these innovations. The <|retrieval|> token is inserted to indicate where additional technical information might be needed.",
"Ideal Knowledge": "Detailed explanations of the single-point urban interchange (SPUI) and the innovative pavement marking technique, including their benefits and how they compare to traditional methods.",
"Key Factual Information": "the first single @-@ point urban interchange ( SPUI ; / ˈspuːiː / ) in Michigan , and a new technique to apply the pavement markings , embedding them into the concrete to reduce the chance of a snowplow scraping them off",
"Prefix reformulated": "What were the two firsts in the freeway project?",
"Prefix-intent": "The project was built with two firsts :",
"Verification": "The KFI directly provides the two innovative features of the project as prompted by the PI."
},
{
"Explanation": "The PI sets up the context for a public event, 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 the 'Southbelt Shuffle' event, including its purpose, activities, attendance, and any similar events held for other freeway openings.",
"Key Factual Information": "walk or bike the South Beltline in an open @-@ house event called the \" Southbelt Shuffle \"",
"Prefix reformulated": "What did MDOT allow the public to do before the freeway opened?",
"Prefix-intent": "In advance of the opening of the freeway to traffic , the Michigan Department of Transportation ( MDOT ) allowed the public to",
"Verification": "The KFI directly describes the public event organized by MDOT as prompted by the PI."
}
]
} |
M @-@ 6 starts at exit 64 on I @-@ 196 in Ottawa County near Hudsonville . The freeway runs southeast from the interchange through the rural Georgetown and Jamestown townships toward the county line . Through this area , MDOT traffic surveys measured a traffic count of 27 @,@ 117 vehicles on average per day , the lowest along the freeway , in 2010 . At Kenowa Avenue , the South Beltline crosses into Byron Township in Kent County . The freeway corridor is bounded on each side by farmland , scattered subdivisions , and small pockets of woodland . Near the Wilson Avenue interchange , M @-@ 6 curves to the northeast around the edge of the Ironwood Golf Course and heads for the interchange with Byron Center Avenue . On either side of the freeway at Byron Center Avenue , there are two hospitals , Metro Health and St. Mary 's Southwest , the former located on the very southern edge of the city of Wyoming . Continuing east , the beltline curves to the southeast and into the cloverleaf interchange complex at US 131 . This " mammoth " interchange stretches over a half mile ( 0 @.@ 8 km ) in one direction and over a mile ( 1 @.@ 6 km ) in the other , encompassing 27 bridges and 18 retaining walls . This makes it the largest freeway interchange in Western Michigan . There are four overpasses which carry M @-@ 6 over the US 131 freeway : two for the main carriageways in each direction and two for the collector @-@ distributor lanes on each side . The auxiliary lanes funnel the traffic using the interchange off the main carriageways to eliminate conflicts between merging streams of traffic . In the eastbound direction , the collector @-@ distributor lane also has access to a ramp for traffic bound for 68th Street , which runs parallel to , and south of , the M @-@ 6 freeway . No such access to 68th Street is provided for westbound M @-@ 6 traffic , although 68th Street traffic can access each direction of M @-@ 6 or US 131 .
| [
{
"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 M @-@ 6 starts at exit 64 on I @-@ 196 in Ottawa County near Hudsonville . The freeway runs southeast from the interchange through the rural Georgetown and Jamestown townships toward the county line . Through this area , MDOT traffic surveys measured a traffic count of 27 @,@ 117 vehicles on average per day , the lowest along the freeway , in 2010 . At Kenowa Avenue , the South Beltline crosses into Byron Township in Kent County . The freeway corridor is bounded on each side by farmland , scattered subdivisions , and small pockets of woodland . Near the Wilson Avenue interchange , M @-@ 6 curves to the northeast around the edge of the Ironwood Golf Course and heads for the interchange with Byron Center Avenue . On either side of the freeway at Byron Center Avenue , there are two hospitals , Metro Health and St. Mary 's Southwest , the former located on the very southern edge of the city of Wyoming . Continuing east , the beltline curves to the southeast and into the cloverleaf interchange complex at US 131 . This \" mammoth \" interchange stretches over a half mile ( 0 @.@ 8 km ) in one direction and over a mile ( 1 @.@ 6 km ) in the other , encompassing 27 bridges and 18 retaining walls . This makes it the largest freeway interchange in Western Michigan . There are four overpasses which carry M @-@ 6 over the US 131 freeway : two for the main carriageways in each direction and two for the collector @-@ distributor lanes on each side . The auxiliary lanes funnel the traffic using the interchange off the main carriageways to eliminate conflicts between merging streams of traffic . In the eastbound direction , the collector @-@ distributor lane also has access to a ramp for traffic bound for 68th Street , which runs parallel to , and south of , the M @-@ 6 freeway . No such access to 68th Street is provided for westbound M @-@ 6 traffic , although 68th Street traffic can access each direction of M @-@ 6 or US 131 . \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|>M @-@ 6 starts at exit 64 on I @-@ 196 in<|end-PI|> <|retrieval|> <|start-KFI|>Ottawa County near Hudsonville<|end-KFI|> . The freeway runs southeast from the interchange through the rural Georgetown and Jamestown townships toward the county line . <|start-PI|>Through this area , MDOT traffic surveys measured a traffic count of<|end-PI|> <|retrieval|> <|start-KFI|>27 @,@ 117 vehicles on average per day , the lowest along the freeway , in 2010<|end-KFI|> . At Kenowa Avenue , the South Beltline crosses into Byron Township in Kent County . The freeway corridor is bounded on each side by farmland , scattered subdivisions , and small pockets of woodland . Near the Wilson Avenue interchange , M @-@ 6 curves to the northeast around the edge of the Ironwood Golf Course and heads for the interchange with Byron Center Avenue . On either side of the freeway at Byron Center Avenue , there are two hospitals , Metro Health and St. Mary 's Southwest , the former located on the very southern edge of the city of Wyoming . Continuing east , the beltline curves to the southeast and into the cloverleaf interchange complex at US 131 . <|start-PI|>This ' mammoth ' interchange stretches over<|end-PI|> <|retrieval|> <|start-KFI|>a half mile ( 0 @.@ 8 km ) in one direction and over a mile ( 1 @.@ 6 km ) in the other , encompassing 27 bridges and 18 retaining walls<|end-KFI|> . This makes it the largest freeway interchange in Western Michigan . There are four overpasses which carry M @-@ 6 over the US 131 freeway : two for the main carriageways in each direction and two for the collector @-@ distributor lanes on each side . The auxiliary lanes funnel the traffic using the interchange off the main carriageways to eliminate conflicts between merging streams of traffic . In the eastbound direction , the collector @-@ distributor lane also has access to a ramp for traffic bound for 68th Street , which runs parallel to , and south of , the M @-@ 6 freeway . No such access to 68th Street is provided for westbound M @-@ 6 traffic , although 68th Street traffic can access each direction of M @-@ 6 or US 131 .",
"none": null,
"results": [
{
"Explanation": "The PI introduces M-6 and sets up a query about its starting point, while the KFI provides the specific location. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "A detailed map of Michigan's highway system, specifically showing the starting point of M-6 at exit 64 on I-196 near Hudsonville in Ottawa County.",
"Key Factual Information": "Ottawa County near Hudsonville",
"Prefix reformulated": "Where does M-6 start?",
"Prefix-intent": "M @-@ 6 starts at exit 64 on I @-@ 196 in",
"Verification": "The KFI directly provides the location information requested in the PI."
},
{
"Explanation": "The PI sets up a query about the traffic count in the area, and the KFI provides the specific number and context. The <|retrieval|> token is inserted to indicate where additional traffic data might be needed.",
"Ideal Knowledge": "MDOT traffic survey data for M-6, showing daily average vehicle counts for different sections of the freeway, with specific emphasis on the 2010 data.",
"Key Factual Information": "27 @,@ 117 vehicles on average per day , the lowest along the freeway , in 2010",
"Prefix reformulated": "What was the traffic count measured by MDOT surveys in the area of M-6?",
"Prefix-intent": "Through this area , MDOT traffic surveys measured a traffic count of",
"Verification": "The KFI directly provides the traffic count information requested in the PI."
},
{
"Explanation": "The PI introduces the 'mammoth' interchange and sets up a query about its size, while the KFI provides specific measurements and structural details. The <|retrieval|> token is inserted to indicate where additional information about the interchange might be needed.",
"Ideal Knowledge": "Detailed specifications of the US 131 interchange, including its dimensions, number of bridges and retaining walls, and comparison to other interchanges in Western Michigan.",
"Key Factual Information": "a half mile ( 0 @.@ 8 km ) in one direction and over a mile ( 1 @.@ 6 km ) in the other , encompassing 27 bridges and 18 retaining walls",
"Prefix reformulated": "How large is the 'mammoth' interchange at US 131?",
"Prefix-intent": "This ' mammoth ' interchange stretches over",
"Verification": "The KFI directly provides the size and structural information about the interchange as prompted by the PI."
}
]
} |
East of US 131 , M @-@ 6 crosses over Division Avenue and enters Gaines Township through a series of sound barrier walls as the freeway ramps merge back into the main lanes . This area had the highest traffic counts in 2011 at 55 @,@ 236 vehicles per day . Near Kalamazoo Avenue , the freeway passes through an area with retail businesses and movie theaters on each side of the interchange ; to the northeast is East Kentwood High School . On the approach to the East Paris Avenue underpass , M @-@ 6 curves first to the northeast and then back to the southeast , passing near one of Steelcase 's office buildings , the pyramid @-@ shaped Corporate Development Center . The freeway begins to curve to the northeast as it crosses into Caledonia Township , with an interchange for M @-@ 37 ( Broadmoor Avenue ) and an overpass for 60th Street . M @-@ 6 curves around the southeast side of the Gerald R. Ford International Airport in Cascade Township . As the South Beltline nears I @-@ 96 , it crosses 48th Street next to the Thornapple Pointe Golf Course . The carriageway splits into ramps for each direction of I @-@ 96 , the ramps to eastbound I @-@ 96 crossing over the Thornapple River in the process . This interchange marks the eastern terminus of the Paul B. Henry South Beltline Freeway . The entire length of the freeway is listed on the National Highway System , a system of roads important to the nation 's economy , defense and mobility .
| [
{
"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 East of US 131 , M @-@ 6 crosses over Division Avenue and enters Gaines Township through a series of sound barrier walls as the freeway ramps merge back into the main lanes . This area had the highest traffic counts in 2011 at 55 @,@ 236 vehicles per day . Near Kalamazoo Avenue , the freeway passes through an area with retail businesses and movie theaters on each side of the interchange ; to the northeast is East Kentwood High School . On the approach to the East Paris Avenue underpass , M @-@ 6 curves first to the northeast and then back to the southeast , passing near one of Steelcase 's office buildings , the pyramid @-@ shaped Corporate Development Center . The freeway begins to curve to the northeast as it crosses into Caledonia Township , with an interchange for M @-@ 37 ( Broadmoor Avenue ) and an overpass for 60th Street . M @-@ 6 curves around the southeast side of the Gerald R. Ford International Airport in Cascade Township . As the South Beltline nears I @-@ 96 , it crosses 48th Street next to the Thornapple Pointe Golf Course . The carriageway splits into ramps for each direction of I @-@ 96 , the ramps to eastbound I @-@ 96 crossing over the Thornapple River in the process . This interchange marks the eastern terminus of the Paul B. Henry South Beltline Freeway . The entire length of the freeway is listed on the National Highway System , a system of roads important to the nation 's economy , defense and mobility . \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": "East of US 131, M-6 crosses over Division Avenue and enters Gaines Township through a series of sound barrier walls as the freeway ramps merge back into the main lanes. <|start-PI|>This area had the highest traffic counts in 2011 at<|end-PI|> <|retrieval|> <|start-KFI|>55,236 vehicles per day<|end-KFI|>. <|start-PI|>Near Kalamazoo Avenue, the freeway passes through an area with<|end-PI|> <|retrieval|> <|start-KFI|>retail businesses and movie theaters on each side of the interchange<|end-KFI|>; to the northeast is East Kentwood High School. On the approach to the East Paris Avenue underpass, M-6 curves first to the northeast and then back to the southeast, passing near one of Steelcase's office buildings, the pyramid-shaped Corporate Development Center. The freeway begins to curve to the northeast as it crosses into Caledonia Township, with an interchange for M-37 (Broadmoor Avenue) and an overpass for 60th Street. <|start-PI|>M-6 curves around the southeast side of<|end-PI|> <|retrieval|> <|start-KFI|>the Gerald R. Ford International Airport in Cascade Township<|end-KFI|>. As the South Beltline nears I-96, it crosses 48th Street next to the Thornapple Pointe Golf Course. The carriageway splits into ramps for each direction of I-96, the ramps to eastbound I-96 crossing over the Thornapple River in the process. This interchange marks the eastern terminus of the Paul B. Henry South Beltline Freeway. <|start-PI|>The entire length of the freeway is listed on<|end-PI|> <|retrieval|> <|start-KFI|>the National Highway System, a system of roads important to the nation's economy, defense and mobility<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI sets up a query about the highest traffic counts in 2011, 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": "A detailed traffic report for the area around M-6 in 2011, including daily vehicle counts for various sections of the freeway.",
"Key Factual Information": "55,236 vehicles per day",
"Prefix reformulated": "What were the highest traffic counts in this area in 2011?",
"Prefix-intent": "This area had the highest traffic counts in 2011 at",
"Verification": "The KFI directly provides the traffic count information requested in the PI."
},
{
"Explanation": "The PI introduces the location and sets up a query about the surrounding area, while the KFI provides specific details about the businesses present. The <|retrieval|> token is inserted to indicate where additional information about the area might be needed.",
"Ideal Knowledge": "A detailed map or description of the commercial areas along M-6 near Kalamazoo Avenue, including specific types of businesses and entertainment venues.",
"Key Factual Information": "retail businesses and movie theaters on each side of the interchange",
"Prefix reformulated": "What type of area does the freeway pass through near Kalamazoo Avenue?",
"Prefix-intent": "Near Kalamazoo Avenue, the freeway passes through an area with",
"Verification": "The KFI directly describes the type of area the freeway passes through, as prompted by the PI."
},
{
"Explanation": "The PI describes the curve of M-6 and sets up a query about what it curves around, while the KFI provides the specific landmark. The <|retrieval|> token is inserted to indicate where additional geographical information might be needed.",
"Ideal Knowledge": "A detailed map of the M-6 route in relation to major landmarks in the area, specifically highlighting its path around the Gerald R. Ford International Airport.",
"Key Factual Information": "the Gerald R. Ford International Airport in Cascade Township",
"Prefix reformulated": "What does M-6 curve around on its southeast side?",
"Prefix-intent": "M-6 curves around the southeast side of",
"Verification": "The KFI directly identifies the landmark that M-6 curves around, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the freeway's classification, and the KFI provides the specific system and its importance. The <|retrieval|> token is inserted to indicate where additional information about highway classifications might be needed.",
"Ideal Knowledge": "Information about the National Highway System, including its criteria for inclusion and the significance of being part of this system.",
"Key Factual Information": "the National Highway System, a system of roads important to the nation's economy, defense and mobility",
"Prefix reformulated": "What system is the entire length of M-6 listed on?",
"Prefix-intent": "The entire length of the freeway is listed on",
"Verification": "The KFI directly identifies the system on which the freeway is listed and provides additional context, as prompted by the PI."
}
]
} |
The right @-@ of @-@ way along M @-@ 6 includes a 9 @-@ mile ( 14 km ) pedestrian path known as the Frederik Meijer Trail . Previously called the M @-@ 6 Trail , it links the Kent Trails west of Byron Center Avenue in Wyoming with the Paul Henry Rail Trail at 60th Street and Wing Avenue by Paris Park in Kentwood .
| [
{
"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 right @-@ of @-@ way along M @-@ 6 includes a 9 @-@ mile ( 14 km ) pedestrian path known as the Frederik Meijer Trail . Previously called the M @-@ 6 Trail , it links the Kent Trails west of Byron Center Avenue in Wyoming with the Paul Henry Rail Trail at 60th Street and Wing Avenue by Paris Park in Kentwood . \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 right-of-way along M-6 includes a 9-mile (14 km) pedestrian path known as<|end-PI|> <|retrieval|> <|start-KFI|>the Frederik Meijer Trail<|end-KFI|>. <|start-PI|>Previously called the M-6 Trail, it links<|end-PI|> <|retrieval|> <|start-KFI|>the Kent Trails west of Byron Center Avenue in Wyoming with the Paul Henry Rail Trail at 60th Street and Wing Avenue by Paris Park in Kentwood<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces the pedestrian path and sets up a query about its name, while the KFI provides the specific name. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "Information about pedestrian paths along M-6, including their names, lengths, and any recent name changes.",
"Key Factual Information": "the Frederik Meijer Trail",
"Prefix reformulated": "What is the name of the 9-mile (14 km) pedestrian path along M-6?",
"Prefix-intent": "The right-of-way along M-6 includes a 9-mile (14 km) pedestrian path known as",
"Verification": "The KFI directly provides the name of the pedestrian path mentioned in the PI."
},
{
"Explanation": "The PI sets up a query about what the trail connects, and the KFI provides specific details about the trail's endpoints. The <|retrieval|> token is inserted to indicate where additional information about the trail connections might be needed.",
"Ideal Knowledge": "Detailed information about the trail network in the area, including the locations of Kent Trails, Paul Henry Rail Trail, and how the Frederik Meijer Trail connects them.",
"Key Factual Information": "the Kent Trails west of Byron Center Avenue in Wyoming with the Paul Henry Rail Trail at 60th Street and Wing Avenue by Paris Park in Kentwood",
"Prefix reformulated": "What does the Frederik Meijer Trail (previously called the M-6 Trail) link?",
"Prefix-intent": "Previously called the M-6 Trail, it links",
"Verification": "The KFI directly provides the information about what the trail links, as prompted by the PI."
}
]
} |
The first appearance of M @-@ 6 was in 1926 as a two @-@ mile ( 3 @.@ 2 km ) road in Keweenaw County in the Upper Peninsula . The highway ran from US 41 at Phoenix to north of Eagle River . The Michigan State Highway Department redesignated the highway as M @-@ 111 in 1938 , and it was redesignated two years later to become a part of the route of M @-@ 26 .
| [
{
"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 appearance of M @-@ 6 was in 1926 as a two @-@ mile ( 3 @.@ 2 km ) road in Keweenaw County in the Upper Peninsula . The highway ran from US 41 at Phoenix to north of Eagle River . The Michigan State Highway Department redesignated the highway as M @-@ 111 in 1938 , and it was redesignated two years later to become a part of the route of M @-@ 26 . \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 first appearance of M @-@ 6 was in<|end-PI|> <|retrieval|> <|start-KFI|>1926 as a two @-@ mile ( 3 @.@ 2 km ) road in Keweenaw County in the Upper Peninsula<|end-KFI|>. <|start-PI|>The highway ran from<|end-PI|> <|retrieval|> <|start-KFI|>US 41 at Phoenix to north of Eagle River<|end-KFI|>. <|start-PI|>The Michigan State Highway Department redesignated the highway as<|end-PI|> <|retrieval|> <|start-KFI|>M @-@ 111 in 1938<|end-KFI|>, and <|start-PI|>it was redesignated two years later to become a part of the route of<|end-PI|> <|retrieval|> <|start-KFI|>M @-@ 26<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI sets up a query about when M-6 first appeared, and the KFI provides the specific year and details. The <|retrieval|> token is inserted to indicate where additional historical information might be needed.",
"Ideal Knowledge": "A chronological list of Michigan highway designations, focusing on the initial appearance of M-6 in 1926, including its original length and location.",
"Key Factual Information": "1926 as a two @-@ mile ( 3 @.@ 2 km ) road in Keweenaw County in the Upper Peninsula",
"Prefix reformulated": "When was the first appearance of M-6?",
"Prefix-intent": "The first appearance of M @-@ 6 was in",
"Verification": "The KFI directly provides the year and details of M-6's first appearance, as prompted by the PI."
},
{
"Explanation": "The PI introduces the highway's route, and the KFI provides the specific start and end points. The <|retrieval|> token is inserted to indicate where additional geographical information might be needed.",
"Ideal Knowledge": "A detailed map or description of the original M-6 route in Keweenaw County, Upper Peninsula, showing its start at US 41 in Phoenix and its endpoint north of Eagle River.",
"Key Factual Information": "US 41 at Phoenix to north of Eagle River",
"Prefix reformulated": "What was the route of the original M-6 highway?",
"Prefix-intent": "The highway ran from",
"Verification": "The KFI directly describes the route of the highway, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the highway's redesignation, and the KFI provides the new designation and year. The <|retrieval|> token is inserted to indicate where additional historical information about highway redesignations might be needed.",
"Ideal Knowledge": "A timeline of Michigan highway redesignations, focusing on the change from M-6 to M-111 in 1938, including any reasons for the change.",
"Key Factual Information": "M @-@ 111 in 1938",
"Prefix reformulated": "What did the Michigan State Highway Department redesignate the M-6 highway as?",
"Prefix-intent": "The Michigan State Highway Department redesignated the highway as",
"Verification": "The KFI directly provides the new designation and year, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about a subsequent redesignation, and the KFI provides the new route number. The <|retrieval|> token is inserted to indicate where additional information about this final change might be needed.",
"Ideal Knowledge": "Information on the 1940 redesignation of M-111, explaining its incorporation into the M-26 route and any changes to the overall highway system in the area.",
"Key Factual Information": "M @-@ 26",
"Prefix reformulated": "What did the M-111 highway become part of two years after its initial redesignation?",
"Prefix-intent": "it was redesignated two years later to become a part of the route of",
"Verification": "The KFI directly provides the new designation, as prompted by the PI."
}
]
} |
In the late 1970s , during the second phase of construction of the I @-@ 696 ( Walter P. Reuther Freeway ) in Metro Detroit , lobbying efforts and lawsuits attempted to block construction of the central section . If successful , the efforts would have left the freeway with a gap in the middle between the first ( western ) and second ( eastern ) phases of construction . During this time , MDOT assigned M @-@ 6 to the eastern section of the freeway under construction . Signs were erected along the service roads that followed 11 Mile Road to connect the already built stack interchange at I @-@ 75 east to I @-@ 94 . By the time the eastern freeway segment was completed in 1979 , the signage for M @-@ 6 was removed and replaced with I @-@ 696 signage , leaving an eight @-@ mile ( 13 km ) gap in the I @-@ 696 freeway until completion of the central section in 1989 .
| [
{
"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 the late 1970s , during the second phase of construction of the I @-@ 696 ( Walter P. Reuther Freeway ) in Metro Detroit , lobbying efforts and lawsuits attempted to block construction of the central section . If successful , the efforts would have left the freeway with a gap in the middle between the first ( western ) and second ( eastern ) phases of construction . During this time , MDOT assigned M @-@ 6 to the eastern section of the freeway under construction . Signs were erected along the service roads that followed 11 Mile Road to connect the already built stack interchange at I @-@ 75 east to I @-@ 94 . By the time the eastern freeway segment was completed in 1979 , the signage for M @-@ 6 was removed and replaced with I @-@ 696 signage , leaving an eight @-@ mile ( 13 km ) gap in the I @-@ 696 freeway until completion of the central section in 1989 . \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 South Beltline Freeway near Grand Rapids was a project that took about 32 years to complete . The idea dates back to the 1940s , but serious proposals were not made until the 1960s . The 1955 planning map for the Grand Rapids area Interstate Highways included a freeway roughly along the M @-@ 6 corridor before I @-@ 96 and I @-@ 196 were shifted north and east to their current locations . An increase in the state gas tax was approved in 1972 with the goal to finance local road projects in the state , including the South Beltline . The project was anticipated to cost $ 30 – 100 million ( equivalent to $ 214 – 713 million in 2015 ) in June 1975 with an expected groundbreaking in 1982 – 85 . The highway was studied in January 1981 for $ 144 @,@ 000 ( equivalent to $ 575 @,@ 000 in 2015 ) . The choice of consultants on the project was controversial ; local planners felt that MDOT picked BKI Inc. only because they used a minority @-@ owned subcontractor and not because they would be qualified for the assignment .
| [
{
"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 South Beltline Freeway near Grand Rapids was a project that took about 32 years to complete . The idea dates back to the 1940s , but serious proposals were not made until the 1960s . The 1955 planning map for the Grand Rapids area Interstate Highways included a freeway roughly along the M @-@ 6 corridor before I @-@ 96 and I @-@ 196 were shifted north and east to their current locations . An increase in the state gas tax was approved in 1972 with the goal to finance local road projects in the state , including the South Beltline . The project was anticipated to cost $ 30 – 100 million ( equivalent to $ 214 – 713 million in 2015 ) in June 1975 with an expected groundbreaking in 1982 – 85 . The highway was studied in January 1981 for $ 144 @,@ 000 ( equivalent to $ 575 @,@ 000 in 2015 ) . The choice of consultants on the project was controversial ; local planners felt that MDOT picked BKI Inc. only because they used a minority @-@ owned subcontractor and not because they would be qualified for the assignment . \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 South Beltline Freeway near Grand Rapids was a project that took about<|end-PI|> <|retrieval|> <|start-KFI|>32 years to complete<|end-KFI|>. The idea dates back to the 1940s, but serious proposals were not made until the 1960s. <|start-PI|>The 1955 planning map for the Grand Rapids area Interstate Highways included a freeway roughly along<|end-PI|> <|retrieval|> <|start-KFI|>the M-6 corridor before I-96 and I-196 were shifted north and east to their current locations<|end-KFI|>. <|start-PI|>An increase in the state gas tax was approved in 1972 with the goal to<|end-PI|> <|retrieval|> <|start-KFI|>finance local road projects in the state, including the South Beltline<|end-KFI|>. <|start-PI|>The project was anticipated to cost<|end-PI|> <|retrieval|> <|start-KFI|>$ 30 – 100 million ( equivalent to $ 214 – 713 million in 2015 ) in June 1975 with an expected groundbreaking in 1982 – 85<|end-KFI|>. <|start-PI|>The highway was studied in January 1981 for<|end-PI|> <|retrieval|> <|start-KFI|>$ 144,000 ( equivalent to $ 575,000 in 2015 )<|end-KFI|>. The choice of consultants on the project was controversial; local planners felt that MDOT picked BKI Inc. only because they used a minority-owned subcontractor and not because they would be qualified for the assignment.",
"none": null,
"results": [
{
"Explanation": "The PI introduces the South Beltline Freeway project and sets up a query about its duration, while the KFI provides the specific timeframe. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "A timeline of the South Beltline Freeway project near Grand Rapids, detailing its inception, major milestones, and completion date, spanning a period of 32 years.",
"Key Factual Information": "32 years to complete",
"Prefix reformulated": "How long did it take to complete the South Beltline Freeway near Grand Rapids?",
"Prefix-intent": "The South Beltline Freeway near Grand Rapids was a project that took about",
"Verification": "The KFI directly provides the duration of the project as requested in the PI."
},
{
"Explanation": "The PI sets up a query about the 1955 planning map, and the KFI provides the specific corridor and additional context. The <|retrieval|> token is inserted to indicate where additional information about historical highway planning might be needed.",
"Ideal Knowledge": "Detailed information about the 1955 planning map for Grand Rapids area Interstate Highways, including the original proposed routes and subsequent changes, specifically mentioning the M-6 corridor and the relocation of I-96 and I-196.",
"Key Factual Information": "the M-6 corridor before I-96 and I-196 were shifted north and east to their current locations",
"Prefix reformulated": "What corridor was included in the 1955 planning map for Grand Rapids area Interstate Highways?",
"Prefix-intent": "The 1955 planning map for the Grand Rapids area Interstate Highways included a freeway roughly along",
"Verification": "The KFI directly provides the specific corridor mentioned in the 1955 planning map, as prompted by the PI."
},
{
"Explanation": "The PI introduces the 1972 gas tax increase and sets up a query about its purpose, while the KFI provides the specific goal. The <|retrieval|> token is inserted to indicate where additional information about the tax increase and its intended use might be needed.",
"Ideal Knowledge": "Information on the 1972 Michigan state gas tax increase, including its purpose, the amount of the increase, and specific projects it was intended to fund, with emphasis on the South Beltline project.",
"Key Factual Information": "finance local road projects in the state, including the South Beltline",
"Prefix reformulated": "What was the goal of the state gas tax increase approved in 1972?",
"Prefix-intent": "An increase in the state gas tax was approved in 1972 with the goal to",
"Verification": "The KFI directly provides the purpose of the gas tax increase, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the anticipated cost of the project, and the KFI provides the specific cost range, adjusted figures, and timeline. The <|retrieval|> token is inserted to indicate where additional information about project cost estimates might be needed.",
"Ideal Knowledge": "Detailed cost estimates for the South Beltline Freeway project from June 1975, including the original projected range and its equivalent in 2015 dollars, along with the expected groundbreaking timeline.",
"Key Factual Information": "$ 30 – 100 million ( equivalent to $ 214 – 713 million in 2015 ) in June 1975 with an expected groundbreaking in 1982 – 85",
"Prefix reformulated": "What was the anticipated cost of the South Beltline Freeway project?",
"Prefix-intent": "The project was anticipated to cost",
"Verification": "The KFI directly provides the anticipated cost range and timeline, as prompted by the PI."
},
{
"Explanation": "The PI introduces a study conducted in January 1981 and sets up a query about its cost, while the KFI provides the specific amount and its adjusted value. The <|retrieval|> token is inserted to indicate where additional information about the study might be needed.",
"Ideal Knowledge": "Information on the January 1981 study of the South Beltline Freeway, including its cost, scope, and findings, with the original cost and its equivalent value in 2015 dollars.",
"Key Factual Information": "$ 144,000 ( equivalent to $ 575,000 in 2015 )",
"Prefix reformulated": "How much did the study of the South Beltline Freeway cost in January 1981?",
"Prefix-intent": "The highway was studied in January 1981 for",
"Verification": "The KFI directly provides the cost of the study and its equivalent value, as prompted by the PI."
}
]
} |
As this study was initiated , the route for the proposed freeway was located between 60th and 68th streets with a western end in Hudsonville and an eastern end in Lowell Township . The consultants were asked to study a full freeway and a limited access boulevard design . One final option was a " no @-@ build " alternative ; under this option , existing roads would be upgraded but no new roads would be built . The City of Grand Rapids opposed the freeway while the suburbs and townships south and west of the city supported it . City officials were concerned about the impact to commercial and industrial business in Grand Rapids . Hudsonville 's city manager favored the proposal as a benefit to local vegetable producers who shipped produce to Detroit or Cleveland . Other supporters , such as the Georgetown Township supervisor , were concerned that delays in starting the project could increase costs . State and local officials expected the freeway in January 1981 to cost between $ 40 – 100 million ( equivalent to $ 160 – 400 million in 2015 ) . The road was to be started no sooner than 1990 .
| [
{
"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 this study was initiated , the route for the proposed freeway was located between 60th and 68th streets with a western end in Hudsonville and an eastern end in Lowell Township . The consultants were asked to study a full freeway and a limited access boulevard design . One final option was a \" no @-@ build \" alternative ; under this option , existing roads would be upgraded but no new roads would be built . The City of Grand Rapids opposed the freeway while the suburbs and townships south and west of the city supported it . City officials were concerned about the impact to commercial and industrial business in Grand Rapids . Hudsonville 's city manager favored the proposal as a benefit to local vegetable producers who shipped produce to Detroit or Cleveland . Other supporters , such as the Georgetown Township supervisor , were concerned that delays in starting the project could increase costs . State and local officials expected the freeway in January 1981 to cost between $ 40 – 100 million ( equivalent to $ 160 – 400 million in 2015 ) . The road was to be started no sooner than 1990 . \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 boundaries for the highway corridor were determined by the consultants in April 1982 , running between 60th and 84th streets , " dipping like a hammock beneath the cities of Kentwood and Wyoming " . The results of the study by BKI were criticized by local planners in May 1982 , who called the study " shabby and unprofessional work " , and asked the state to fire the consulting firm . Local residents distributed 2 @,@ 000 fliers to their neighbors in opposition to the freeway . The South Belt Local Advisory Board criticized BKI 's 110 @-@ page study report as " filled with errors " ; the consultants ' earlier 26 @-@ page paper had been rejected by the board and MDOT as " unusable " . An editorial in the The Grand Rapids Press stated that the study did not help advance the project in the area , instead opening the proposed freeway up to new controversies . Doubts about the state 's budget in 1982 to build the roadway combined with issues over the consultants and their study results .
| [
{
"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 boundaries for the highway corridor were determined by the consultants in April 1982 , running between 60th and 84th streets , \" dipping like a hammock beneath the cities of Kentwood and Wyoming \" . The results of the study by BKI were criticized by local planners in May 1982 , who called the study \" shabby and unprofessional work \" , and asked the state to fire the consulting firm . Local residents distributed 2 @,@ 000 fliers to their neighbors in opposition to the freeway . The South Belt Local Advisory Board criticized BKI 's 110 @-@ page study report as \" filled with errors \" ; the consultants ' earlier 26 @-@ page paper had been rejected by the board and MDOT as \" unusable \" . An editorial in the The Grand Rapids Press stated that the study did not help advance the project in the area , instead opening the proposed freeway up to new controversies . Doubts about the state 's budget in 1982 to build the roadway combined with issues over the consultants and their study results . \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 second citizens group , the South Belt Citizens Committee , was formed in July 1982 to gain additional public information on the project and supplement the work of the other groups , including the South Belt Local Advisory Board . BKI was fired as consultant on the project by MDOT on September 9 , 1982 . The switch to a new consultant delayed further study because of the timetable to take bids and interview the candidates . In the interim , work was shifted to local and state planners until a new consultant could be retained in an effort to minimize the delays involved . These local projects were focused on updating the information and maps from BKI 's study and refining the scope of the highway 's corridor .
| [
{
"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 second citizens group , the South Belt Citizens Committee , was formed in July 1982 to gain additional public information on the project and supplement the work of the other groups , including the South Belt Local Advisory Board . BKI was fired as consultant on the project by MDOT on September 9 , 1982 . The switch to a new consultant delayed further study because of the timetable to take bids and interview the candidates . In the interim , work was shifted to local and state planners until a new consultant could be retained in an effort to minimize the delays involved . These local projects were focused on updating the information and maps from BKI 's study and refining the scope of the highway 's corridor . \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|>A second citizens group , the South Belt Citizens Committee , was formed in July 1982 to<|end-PI|> <|retrieval|> <|start-KFI|>gain additional public information on the project and supplement the work of the other groups , including the South Belt Local Advisory Board<|end-KFI|> . <|start-PI|>BKI was fired as consultant on the project by MDOT on<|end-PI|> <|retrieval|> <|start-KFI|>September 9 , 1982<|end-KFI|> . <|start-PI|>The switch to a new consultant delayed further study because of<|end-PI|> <|retrieval|> <|start-KFI|>the timetable to take bids and interview the candidates<|end-KFI|> . <|start-PI|>In the interim , work was shifted to local and state planners until a new consultant could be retained in an effort to<|end-PI|> <|retrieval|> <|start-KFI|>minimize the delays involved<|end-KFI|> . <|start-PI|>These local projects were focused on<|end-PI|> <|retrieval|> <|start-KFI|>updating the information and maps from BKI 's study and refining the scope of the highway 's corridor<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the formation of a new citizens group and sets up a query about its purpose, while the KFI provides the specific reasons for its formation. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "Detailed information about the formation of citizen groups related to the South Belt project, including their purposes, dates of establishment, and relationships with other groups.",
"Key Factual Information": "gain additional public information on the project and supplement the work of the other groups , including the South Belt Local Advisory Board",
"Prefix reformulated": "What was the purpose of forming the South Belt Citizens Committee in July 1982?",
"Prefix-intent": "A second citizens group , the South Belt Citizens Committee , was formed in July 1982 to",
"Verification": "The KFI directly provides the purpose of forming the committee, as requested in the PI."
},
{
"Explanation": "The PI sets up a query about when BKI was fired, and the KFI provides the specific date. The <|retrieval|> token is inserted to indicate where additional information about the project's timeline might be needed.",
"Ideal Knowledge": "A timeline of events related to the South Belt project, specifically focusing on the involvement and termination of consultants, including BKI.",
"Key Factual Information": "September 9 , 1982",
"Prefix reformulated": "When was BKI fired as consultant on the project by MDOT?",
"Prefix-intent": "BKI was fired as consultant on the project by MDOT on",
"Verification": "The KFI directly provides the date when BKI was fired, as prompted by the PI."
},
{
"Explanation": "The PI introduces the delay caused by switching consultants and sets up a query about the reason, while the KFI provides the specific cause of the delay. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the explanation.",
"Ideal Knowledge": "Information on the process of hiring consultants for large-scale projects, including typical timelines for bidding and interviewing, and how this process can impact project schedules.",
"Key Factual Information": "the timetable to take bids and interview the candidates",
"Prefix reformulated": "Why did the switch to a new consultant delay further study on the South Belt project?",
"Prefix-intent": "The switch to a new consultant delayed further study because of",
"Verification": "The KFI directly explains the reason for the delay, as prompted by the PI."
},
{
"Explanation": "The PI describes the shift in work and sets up a query about its purpose, while the KFI provides the specific reason for this action. The <|retrieval|> token is inserted to indicate where additional information about project management strategies might be needed.",
"Ideal Knowledge": "Information on project management strategies for large-scale infrastructure projects, particularly focusing on how work is redistributed when there are changes in key personnel or consultants.",
"Key Factual Information": "minimize the delays involved",
"Prefix reformulated": "Why was work shifted to local and state planners during the interim period of the South Belt project?",
"Prefix-intent": "In the interim , work was shifted to local and state planners until a new consultant could be retained in an effort to",
"Verification": "The KFI directly provides the reason for shifting work to local and state planners, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the focus of the local projects, and the KFI provides the specific tasks they were working on. The <|retrieval|> token is inserted to indicate where additional information about project planning and refinement might be needed.",
"Ideal Knowledge": "Detailed information about the specific tasks involved in updating and refining highway project plans, including the importance of accurate maps and corridor definitions in infrastructure planning.",
"Key Factual Information": "updating the information and maps from BKI 's study and refining the scope of the highway 's corridor",
"Prefix reformulated": "What were the local projects related to the South Belt project focused on?",
"Prefix-intent": "These local projects were focused on",
"Verification": "The KFI directly describes the focus of the local projects, as prompted by the PI."
}
]
} |
Gaines , Cascade , and Caledonia townships and the city of Kentwood circulated a survey amongst their communities ' planning commissions and elected boards in 1982 . The survey showed an inconclusive preference for a limited @-@ access highway in what was termed a " gut level reaction " to the proposed roadway . The South Belt Citizens Committee actively started to oppose the roadway during the fall of 1982 , pressing local candidates for political office to take positions on the project . The group called the roadway a " holocaust " in its mailings to the candidates . Future Lieutenant Governor Dick Posthumus , then a member of the Michigan State Senate , called for an end to the studies in 1983 ; he would later reverse and become one of the project 's biggest supporters .
| [
{
"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 Gaines , Cascade , and Caledonia townships and the city of Kentwood circulated a survey amongst their communities ' planning commissions and elected boards in 1982 . The survey showed an inconclusive preference for a limited @-@ access highway in what was termed a \" gut level reaction \" to the proposed roadway . The South Belt Citizens Committee actively started to oppose the roadway during the fall of 1982 , pressing local candidates for political office to take positions on the project . The group called the roadway a \" holocaust \" in its mailings to the candidates . Future Lieutenant Governor Dick Posthumus , then a member of the Michigan State Senate , called for an end to the studies in 1983 ; he would later reverse and become one of the project 's biggest supporters . \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
} |
MDOT hired a new consultant , Schimpeler / Coradino Associates , in 1984 to study the proposed freeway . The consultant recommended the freeway in March 1985 . The South Beltline was included in the ten @-@ year highway plan in 1986 . By 1989 , the state wanted to set the route in 1991 with construction starting in 1993 . The freeway was studied as a possible toll road in June 1991 after three alternative routes were proposed the previous year . That September , the final route was set with a projected start date in 1997 . The toll road concept was revived in September 1995 to offset the failure of a proposed gas tax increase . As a cost @-@ saving measure , the number of interchanges was reduced to four from eight in June 1996 . When a gas tax increase was passed in 1997 , Governor John Engler promised at least seven access points for the freeway . The South Beltline was touted as " [ cutting ] travel time around Grand Rapids virtually in half " .
| [
{
"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 MDOT hired a new consultant , Schimpeler / Coradino Associates , in 1984 to study the proposed freeway . The consultant recommended the freeway in March 1985 . The South Beltline was included in the ten @-@ year highway plan in 1986 . By 1989 , the state wanted to set the route in 1991 with construction starting in 1993 . The freeway was studied as a possible toll road in June 1991 after three alternative routes were proposed the previous year . That September , the final route was set with a projected start date in 1997 . The toll road concept was revived in September 1995 to offset the failure of a proposed gas tax increase . As a cost @-@ saving measure , the number of interchanges was reduced to four from eight in June 1996 . When a gas tax increase was passed in 1997 , Governor John Engler promised at least seven access points for the freeway . The South Beltline was touted as \" [ cutting ] travel time around Grand Rapids virtually in half \" . \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
} |
Proposals for the South Beltline Freeway were nearly 25 years old by the time initial construction was started in 1997 . The Michigan State Legislature named the South Beltline around the same time for the Congressman Paul B. Henry , who died in office in 1993 , serving in Gerald Ford 's old US House seat . The cost of the construction of new roads like the South Beltline was a campaign issue when Engler ran for re @-@ election against Geoffrey Feiger in 1998 . The entire freeway was projected to open by 2008 , with the first phase opening in 2002 . MDOT gave the South Beltline its numerical designation on the July 1999 edition of the state map , marking M @-@ 6 for the first time as a dotted line , to denote it was " under construction " . The legislature approved Engler 's " Build Michigan III " program in 2000 ; the plan accelerated road projects in the state . The capital outlay for the year was $ 82 million ( equivalent to $ 132 million in 2015 ) . Condemnation proceedings were initiated in the Kent County Circuit Court in 1999 to clear the way for the acquisitions . Land that contained homes , farms , trailer parks , and businesses was purchased by MDOT to acquire the right @-@ of @-@ way needed for the freeway . The land needed measured 360 feet ( 110 m ) wide and 20 miles ( 32 km ) long . Land acquisitions for the South Beltline Freeway were completed in July 2001 . Construction started later in the fall of 2001 on the second and third phases of the project .
| [
{
"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 Proposals for the South Beltline Freeway were nearly 25 years old by the time initial construction was started in 1997 . The Michigan State Legislature named the South Beltline around the same time for the Congressman Paul B. Henry , who died in office in 1993 , serving in Gerald Ford 's old US House seat . The cost of the construction of new roads like the South Beltline was a campaign issue when Engler ran for re @-@ election against Geoffrey Feiger in 1998 . The entire freeway was projected to open by 2008 , with the first phase opening in 2002 . MDOT gave the South Beltline its numerical designation on the July 1999 edition of the state map , marking M @-@ 6 for the first time as a dotted line , to denote it was \" under construction \" . The legislature approved Engler 's \" Build Michigan III \" program in 2000 ; the plan accelerated road projects in the state . The capital outlay for the year was $ 82 million ( equivalent to $ 132 million in 2015 ) . Condemnation proceedings were initiated in the Kent County Circuit Court in 1999 to clear the way for the acquisitions . Land that contained homes , farms , trailer parks , and businesses was purchased by MDOT to acquire the right @-@ of @-@ way needed for the freeway . The land needed measured 360 feet ( 110 m ) wide and 20 miles ( 32 km ) long . Land acquisitions for the South Beltline Freeway were completed in July 2001 . Construction started later in the fall of 2001 on the second and third phases of the project . \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 leg of the South Beltline Freeway , located between M @-@ 37 ( Broadmoor Avenue ) and I @-@ 96 , was finished six months early . Dry summer weather allowed the roadbed contractors to finish their portion of the five @-@ mile ( 8 @.@ 0 km ) section of the freeway early , earning them a $ 300 @,@ 000 bonus ( equivalent to $ 449 @,@ 000 in 2015 ) . The overpasses for the remaining sections of the freeway were completed while the first phase was under construction , leaving the interchanges at US 131 and I @-@ 196 and the connecting roadbed to be completed at that time . The first section was paved in asphalt after MDOT reversed the decision to pave the whole freeway in concrete . That stretch of freeway opened to traffic on November 20 , 2001 . The state kept the overall project in an accelerated status headed into the next phases in 2002 .
| [
{
"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 leg of the South Beltline Freeway , located between M @-@ 37 ( Broadmoor Avenue ) and I @-@ 96 , was finished six months early . Dry summer weather allowed the roadbed contractors to finish their portion of the five @-@ mile ( 8 @.@ 0 km ) section of the freeway early , earning them a $ 300 @,@ 000 bonus ( equivalent to $ 449 @,@ 000 in 2015 ) . The overpasses for the remaining sections of the freeway were completed while the first phase was under construction , leaving the interchanges at US 131 and I @-@ 196 and the connecting roadbed to be completed at that time . The first section was paved in asphalt after MDOT reversed the decision to pave the whole freeway in concrete . That stretch of freeway opened to traffic on November 20 , 2001 . The state kept the overall project in an accelerated status headed into the next phases in 2002 . \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
} |
Construction of the remaining phases between US 131 and M @-@ 37 and between I @-@ 196 and US 131 was started on April 1 , 2002 . Area roads that crossed the path of the new freeway were closed to traffic with posted detours so that work could begin on the roadbed for the freeway . The last major project for the freeway was to replace bridge beams in the overpasses from westbound I @-@ 196 to eastbound M @-@ 6 . Design flaws were found in 2002 in the size of the beams in the bridges over eastbound I @-@ 196 and the ramp from westbound M @-@ 6 to westbound I @-@ 196 . The replacement was originally supposed to close traffic along I @-@ 196 over a weekend in 2004 , but kept a lane closed for a full week , backing up traffic on the Interstate for two miles ( 3 @.@ 2 km ) ; completion of the work was delayed when human error caused a shortage of nuts and bolts .
| [
{
"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 Construction of the remaining phases between US 131 and M @-@ 37 and between I @-@ 196 and US 131 was started on April 1 , 2002 . Area roads that crossed the path of the new freeway were closed to traffic with posted detours so that work could begin on the roadbed for the freeway . The last major project for the freeway was to replace bridge beams in the overpasses from westbound I @-@ 196 to eastbound M @-@ 6 . Design flaws were found in 2002 in the size of the beams in the bridges over eastbound I @-@ 196 and the ramp from westbound M @-@ 6 to westbound I @-@ 196 . The replacement was originally supposed to close traffic along I @-@ 196 over a weekend in 2004 , but kept a lane closed for a full week , backing up traffic on the Interstate for two miles ( 3 @.@ 2 km ) ; completion of the work was delayed when human error caused a shortage of nuts and bolts . \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|>Construction of the remaining phases between US 131 and M @-@ 37 and between I @-@ 196 and US 131 was started on<|end-PI|> <|retrieval|> <|start-KFI|>April 1 , 2002<|end-KFI|> . Area roads that crossed the path of the new freeway were closed to traffic with posted detours so that work could begin on the roadbed for the freeway . <|start-PI|>The last major project for the freeway was to<|end-PI|> <|retrieval|> <|start-KFI|>replace bridge beams in the overpasses from westbound I @-@ 196 to eastbound M @-@ 6<|end-KFI|> . <|start-PI|>Design flaws were found in 2002 in<|end-PI|> <|retrieval|> <|start-KFI|>the size of the beams in the bridges over eastbound I @-@ 196 and the ramp from westbound M @-@ 6 to westbound I @-@ 196<|end-KFI|> . <|start-PI|>The replacement was originally supposed to close traffic along I @-@ 196 over a weekend in 2004 , but<|end-PI|> <|retrieval|> <|start-KFI|>kept a lane closed for a full week , backing up traffic on the Interstate for two miles ( 3 @.@ 2 km )<|end-KFI|> ; completion of the work was delayed when human error caused a shortage of nuts and bolts .",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context for the construction start date, and the KFI provides the specific date. The <|retrieval|> token is inserted to indicate where additional information about the construction timeline might be needed.",
"Ideal Knowledge": "A timeline of the construction phases for the freeway project, specifically highlighting the start date of April 1, 2002 for the mentioned sections.",
"Key Factual Information": "April 1 , 2002",
"Prefix reformulated": "When did the construction of the remaining phases of the freeway between US 131 and M-37 and between I-196 and US 131 start?",
"Prefix-intent": "Construction of the remaining phases between US 131 and M @-@ 37 and between I @-@ 196 and US 131 was started on",
"Verification": "The KFI directly provides the date requested in the PI."
},
{
"Explanation": "The PI introduces the last major project, and the KFI provides the specific details of that project. The <|retrieval|> token is inserted to indicate where additional information about the project might be needed.",
"Ideal Knowledge": "Detailed information about the final stages of the freeway construction project, focusing on the replacement of bridge beams in specific overpasses.",
"Key Factual Information": "replace bridge beams in the overpasses from westbound I @-@ 196 to eastbound M @-@ 6",
"Prefix reformulated": "What was the last major project for the freeway?",
"Prefix-intent": "The last major project for the freeway was to",
"Verification": "The KFI directly describes the last major project mentioned in the PI."
},
{
"Explanation": "The PI sets up the context for the design flaws, and the KFI provides the specific details of those flaws. The <|retrieval|> token is inserted to indicate where additional information about the design issues might be needed.",
"Ideal Knowledge": "A report detailing the design flaws discovered in 2002, specifically related to beam sizes in certain bridges and ramps of the freeway project.",
"Key Factual Information": "the size of the beams in the bridges over eastbound I @-@ 196 and the ramp from westbound M @-@ 6 to westbound I @-@ 196",
"Prefix reformulated": "What design flaws were found in 2002 in the freeway project?",
"Prefix-intent": "Design flaws were found in 2002 in",
"Verification": "The KFI directly describes the design flaws mentioned in the PI."
},
{
"Explanation": "The PI sets up the original plan for the replacement work, and the KFI provides information on how it actually unfolded. The <|retrieval|> token is inserted to indicate where additional information about the project's execution might be needed.",
"Ideal Knowledge": "A comparison between the planned and actual timeline for the replacement work on I-196 in 2004, including details on traffic impacts and delays.",
"Key Factual Information": "kept a lane closed for a full week , backing up traffic on the Interstate for two miles ( 3 @.@ 2 km )",
"Prefix reformulated": "How did the actual replacement work differ from the original plan for I-196 in 2004?",
"Prefix-intent": "The replacement was originally supposed to close traffic along I @-@ 196 over a weekend in 2004 , but",
"Verification": "The KFI directly describes how the actual replacement work differed from the original plan mentioned in the PI."
}
]
} |
MDOT hosted an open house along the unopened section of M @-@ 6 between Kalamazoo and Byron Center avenues . This event took place on October 2 , 2004 , and was billed as the " Southbelt Shuffle " , allowing the public to walk or bike along the freeway . The event was planned to draw attention to the M @-@ 6 Trail that runs parallel to the freeway . Some event participants brought their horses for the chance to ride on the freeway . The whole freeway was opened to traffic on November 17 , 2004 , after a ribbon cutting ceremony . When opened , reconstruction work was still being completed on overpass bridges at the I @-@ 196 interchange on the west end . The entire project cost $ 700 million ( equivalent to $ 933 million in 2015 ) to complete over the five @-@ year construction period , about $ 35 million / mi ( approximately $ 22 million / km , equivalent to $ 47 million / mi or $ 29 million / km in 2015 ) . When the freeway was opened , it was the first in the state of Michigan to use a SPUI located at the Kalamazoo Avenue exit . All of the bridges and sound barrier walls were painted sienna beige as part of a " color theme " to the freeway . MDOT also used a new technique to recess the pavement markings into the concrete , designed to reduce the likelihood that snowplows would scrape them off . The signs are in miles , but " the entire M @-@ 6 freeway was designed and constructed in metric " , according to MDOT manager Suzette Peplinski . The final ramps opened to traffic on December 9 , 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 MDOT hosted an open house along the unopened section of M @-@ 6 between Kalamazoo and Byron Center avenues . This event took place on October 2 , 2004 , and was billed as the \" Southbelt Shuffle \" , allowing the public to walk or bike along the freeway . The event was planned to draw attention to the M @-@ 6 Trail that runs parallel to the freeway . Some event participants brought their horses for the chance to ride on the freeway . The whole freeway was opened to traffic on November 17 , 2004 , after a ribbon cutting ceremony . When opened , reconstruction work was still being completed on overpass bridges at the I @-@ 196 interchange on the west end . The entire project cost $ 700 million ( equivalent to $ 933 million in 2015 ) to complete over the five @-@ year construction period , about $ 35 million / mi ( approximately $ 22 million / km , equivalent to $ 47 million / mi or $ 29 million / km in 2015 ) . When the freeway was opened , it was the first in the state of Michigan to use a SPUI located at the Kalamazoo Avenue exit . All of the bridges and sound barrier walls were painted sienna beige as part of a \" color theme \" to the freeway . MDOT also used a new technique to recess the pavement markings into the concrete , designed to reduce the likelihood that snowplows would scrape them off . The signs are in miles , but \" the entire M @-@ 6 freeway was designed and constructed in metric \" , according to MDOT manager Suzette Peplinski . The final ramps opened to traffic on December 9 , 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": null,
"none": "",
"results": null
} |
MDOT added the completed M @-@ 6 to the state maps in an updated printing in June 2005 . At the time , the various online mapping services still did not show a complete freeway in eastern Ottawa or southern Kent counties . Services such as Mapquest and Yahoo ! Maps rely on Tele Atlas out of New Hampshire for their mapping information , which , in turn , relies on agencies like MDOT to update their data . MDOT 's map update came out nearly eight months after the initial opening due to its inclusion in a large @-@ scale update to the state highway map .
| [
{
"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 MDOT added the completed M @-@ 6 to the state maps in an updated printing in June 2005 . At the time , the various online mapping services still did not show a complete freeway in eastern Ottawa or southern Kent counties . Services such as Mapquest and Yahoo ! Maps rely on Tele Atlas out of New Hampshire for their mapping information , which , in turn , relies on agencies like MDOT to update their data . MDOT 's map update came out nearly eight months after the initial opening due to its inclusion in a large @-@ scale update to the state highway map . \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 year after the freeway opened , traffic volumes along parallel roads like 44th , 56th and 68th streets dropped 40 – 50 % . At the same time , roads with interchanges along the freeway saw increased traffic . Wilson Avenue experienced a 120 % increase and sections of Byron Center Avenue jumped 100 % in traffic levels a month after M @-@ 6 opened . Property values in the townships surrounding the freeway increased 11 @.@ 3 – 12 @.@ 4 % by 2006 as a result of development attached to the freeway . Local officials credited the freeway for increased access to the area , driving housing starts as residents flocked to the communities for their schools and quality of life . In 2007 , the Metro Health Village , a commercial development centered around the Byron Center Avenue exit and the hospital opened . Described as being similar to a mall with the hospital as a tenant , the village features restaurants , shops , offices and a hotel . Metro Health relocated from Grand Rapids to the location in Wyoming in the face of opposition to planned expansions of their previous location . Since opening , even though the freeway was officially named for Paul Henry , the original South Beltline name is still in use .
| [
{
"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 year after the freeway opened , traffic volumes along parallel roads like 44th , 56th and 68th streets dropped 40 – 50 % . At the same time , roads with interchanges along the freeway saw increased traffic . Wilson Avenue experienced a 120 % increase and sections of Byron Center Avenue jumped 100 % in traffic levels a month after M @-@ 6 opened . Property values in the townships surrounding the freeway increased 11 @.@ 3 – 12 @.@ 4 % by 2006 as a result of development attached to the freeway . Local officials credited the freeway for increased access to the area , driving housing starts as residents flocked to the communities for their schools and quality of life . In 2007 , the Metro Health Village , a commercial development centered around the Byron Center Avenue exit and the hospital opened . Described as being similar to a mall with the hospital as a tenant , the village features restaurants , shops , offices and a hotel . Metro Health relocated from Grand Rapids to the location in Wyoming in the face of opposition to planned expansions of their previous location . Since opening , even though the freeway was officially named for Paul Henry , the original South Beltline name is still in use . \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
} |
Reactions to the new freeway were not all positive . In a special editorial in the The Grand Rapids Press after the freeway opened in 2004 , local resident Curt MacDougall summarized the criticisms of the new freeway . He cited the loss of rural farmland and wetlands as a negative effect of the highway . The editorial also discussed that the freeway does decrease travel times for some residents , but it will mean increased development . That development will mean further urban sprawl , and could spur the creation of additional highways in the area .
| [
{
"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 Reactions to the new freeway were not all positive . In a special editorial in the The Grand Rapids Press after the freeway opened in 2004 , local resident Curt MacDougall summarized the criticisms of the new freeway . He cited the loss of rural farmland and wetlands as a negative effect of the highway . The editorial also discussed that the freeway does decrease travel times for some residents , but it will mean increased development . That development will mean further urban sprawl , and could spur the creation of additional highways in the area . \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": "Reactions to the new freeway were not all positive . <|start-PI|> In a special editorial in the The Grand Rapids Press after the freeway opened in <|end-PI|> <|retrieval|> <|start-KFI|> 2004 <|end-KFI|> , local resident Curt MacDougall summarized the criticisms of the new freeway . <|start-PI|> He cited the loss of rural farmland and wetlands as <|end-PI|> <|retrieval|> <|start-KFI|> a negative effect of the highway <|end-KFI|> . <|start-PI|> The editorial also discussed that the freeway does decrease travel times for some residents , but it will mean <|end-PI|> <|retrieval|> <|start-KFI|> increased development <|end-KFI|> . <|start-PI|> That development will mean further urban sprawl , and could spur <|end-PI|> <|retrieval|> <|start-KFI|> the creation of additional highways in the area <|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context for when the freeway opened, and the KFI provides the specific year. The <|retrieval|> token is inserted to indicate where additional information about the freeway's opening date might be needed.",
"Ideal Knowledge": "A timeline of the freeway's construction and opening, with specific dates and events related to its completion in 2004.",
"Key Factual Information": "2004",
"Prefix reformulated": "When did the new freeway open according to The Grand Rapids Press editorial?",
"Prefix-intent": "Reactions to the new freeway were not all positive . In a special editorial in the The Grand Rapids Press after the freeway opened in",
"Verification": "The KFI directly provides the year mentioned in the PI."
},
{
"Explanation": "The PI introduces the criticism, and the KFI specifies the negative effect. The <|retrieval|> token is inserted to indicate where additional information about the environmental impact might be needed.",
"Ideal Knowledge": "A detailed analysis of the environmental impact of the freeway, including statistics on the loss of rural farmland and wetlands in the affected area.",
"Key Factual Information": "a negative effect of the highway",
"Prefix reformulated": "What did Curt MacDougall cite as a negative effect of the highway?",
"Prefix-intent": "He cited the loss of rural farmland and wetlands as",
"Verification": "The KFI directly states the negative effect mentioned in the PI."
},
{
"Explanation": "The PI sets up the contrast between decreased travel times and another consequence, and the KFI provides that consequence. The <|retrieval|> token is inserted to indicate where additional information about the impact of freeways on development might be needed.",
"Ideal Knowledge": "Studies on the relationship between new highway construction and urban development, with specific examples of how freeways can lead to increased development in surrounding areas.",
"Key Factual Information": "increased development",
"Prefix reformulated": "What will the freeway mean according to the editorial, despite decreasing travel times?",
"Prefix-intent": "The editorial also discussed that the freeway does decrease travel times for some residents , but it will mean",
"Verification": "The KFI directly provides the consequence mentioned in the PI."
},
{
"Explanation": "The PI introduces the concept of urban sprawl and sets up a query about its consequences, while the KFI provides the specific potential outcome. The <|retrieval|> token is inserted to indicate where additional information about the relationship between urban sprawl and highway construction might be needed.",
"Ideal Knowledge": "Research on the cycle of urban sprawl and highway construction, including case studies of areas where new highways led to further development and subsequent highway construction.",
"Key Factual Information": "the creation of additional highways in the area",
"Prefix reformulated": "What could the development resulting from the freeway spur?",
"Prefix-intent": "That development will mean further urban sprawl , and could spur",
"Verification": "The KFI directly states what the development could lead to, as prompted by the PI."
}
]
} |
The M @-@ 6 Trail was constructed in a $ 3 @.@ 5 million project ( equivalent to $ 4 million in 2015 ) that started in 2008 . The goal was to create a 10 @-@ foot @-@ wide ( 3 @.@ 0 m ) path linking the Kent Trails with the Paul Henry Rail Trail . The M @-@ 6 Trail was the brainchild of Gaines Township Supervisor Don Hilton , Sr. He had pushed to have the path included in the original freeway construction and opened with the rest of the South Beltline . The trail project was funded by $ 2 @.@ 9 million ( equivalent to $ 3 @.@ 3 million in 2015 ) in federal grants and $ 300 @,@ 000 ( equivalent to $ 350 @,@ 000 in 2015 ) from the Frederik and Lena Meijer Foundation . The balance came from Kent County and the townships . Work on the trail was completed in November 2008 .
| [
{
"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 M @-@ 6 Trail was constructed in a $ 3 @.@ 5 million project ( equivalent to $ 4 million in 2015 ) that started in 2008 . The goal was to create a 10 @-@ foot @-@ wide ( 3 @.@ 0 m ) path linking the Kent Trails with the Paul Henry Rail Trail . The M @-@ 6 Trail was the brainchild of Gaines Township Supervisor Don Hilton , Sr. He had pushed to have the path included in the original freeway construction and opened with the rest of the South Beltline . The trail project was funded by $ 2 @.@ 9 million ( equivalent to $ 3 @.@ 3 million in 2015 ) in federal grants and $ 300 @,@ 000 ( equivalent to $ 350 @,@ 000 in 2015 ) from the Frederik and Lena Meijer Foundation . The balance came from Kent County and the townships . Work on the trail was completed in November 2008 . \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 M @-@ 6 Trail was constructed in <|start-PI|> a $ 3 @.@ 5 million project ( equivalent to <|end-PI|> <|retrieval|> <|start-KFI|> $ 4 million in 2015 ) <|end-KFI|> that started in 2008 . <|start-PI|> The goal was to create a 10 @-@ foot @-@ wide ( 3 @.@ 0 m ) path linking <|end-PI|> <|retrieval|> <|start-KFI|> the Kent Trails with the Paul Henry Rail Trail <|end-KFI|> . <|start-PI|> The M @-@ 6 Trail was the brainchild of Gaines Township Supervisor <|end-PI|> <|retrieval|> <|start-KFI|> Don Hilton , Sr. <|end-KFI|> He had pushed to have the path included in the original freeway construction and opened with the rest of the South Beltline . <|start-PI|> The trail project was funded by $ 2 @.@ 9 million ( equivalent to <|end-PI|> <|retrieval|> <|start-KFI|> $ 3 @.@ 3 million in 2015 ) in federal grants <|end-KFI|> <|start-PI|> and $ 300 @,@ 000 ( equivalent to <|end-PI|> <|retrieval|> <|start-KFI|> $ 350 @,@ 000 in 2015 ) from the Frederik and Lena Meijer Foundation <|end-KFI|> . The balance came from Kent County and the townships . Work on the trail was completed in November 2008 .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the M-6 Trail construction project and its cost, while the KFI provides the equivalent cost in 2015. The <|retrieval|> token is inserted to indicate where additional information about cost conversion might be needed.",
"Ideal Knowledge": "A detailed breakdown of the M-6 Trail construction project costs, including the original budget and its equivalent value in 2015 dollars, accounting for inflation.",
"Key Factual Information": "$ 4 million in 2015 )",
"Prefix reformulated": "What was the cost of the M-6 Trail construction project in 2015 dollars?",
"Prefix-intent": "The M-6 Trail was constructed in a $ 3.5 million project ( equivalent to",
"Verification": "The KFI directly provides the equivalent cost in 2015 dollars as prompted by the PI."
},
{
"Explanation": "The PI sets up the goal of the M-6 Trail project, and the KFI provides the specific trails it aimed to connect. The <|retrieval|> token is inserted to indicate where additional information about these trails might be needed.",
"Ideal Knowledge": "Information about the Kent Trails and Paul Henry Rail Trail, including their locations, lengths, and significance in the local trail network.",
"Key Factual Information": "the Kent Trails with the Paul Henry Rail Trail",
"Prefix reformulated": "What trails did the M-6 Trail aim to connect?",
"Prefix-intent": "The goal was to create a 10-foot-wide ( 3.0 m ) path linking",
"Verification": "The KFI directly states the trails that the M-6 Trail was designed to connect, as prompted by the PI."
},
{
"Explanation": "The PI introduces the origin of the M-6 Trail idea, and the KFI provides the name of the person responsible. The <|retrieval|> token is inserted to indicate where additional information about Don Hilton, Sr. might be needed.",
"Ideal Knowledge": "Biographical information about Don Hilton, Sr., including his role as Gaines Township Supervisor and his involvement in local infrastructure projects.",
"Key Factual Information": "Don Hilton, Sr.",
"Prefix reformulated": "Who was the person behind the idea of the M-6 Trail?",
"Prefix-intent": "The M-6 Trail was the brainchild of Gaines Township Supervisor",
"Verification": "The KFI directly provides the name of the person who conceived the M-6 Trail idea, as prompted by the PI."
},
{
"Explanation": "The PI introduces the funding source for the trail project, and the KFI provides the specific amount and its 2015 equivalent. The <|retrieval|> token is inserted to indicate where additional information about federal grants might be needed.",
"Ideal Knowledge": "Detailed information about federal grant programs for trail projects, including typical funding amounts and the process for adjusting historical amounts for inflation.",
"Key Factual Information": "$ 3.3 million in 2015 ) in federal grants",
"Prefix reformulated": "What was the value of the federal grants for the M-6 Trail project in 2015 dollars?",
"Prefix-intent": "The trail project was funded by $ 2.9 million ( equivalent to",
"Verification": "The KFI directly provides the equivalent value of the federal grants in 2015 dollars, as prompted by the PI."
},
{
"Explanation": "The PI introduces an additional funding source for the trail project, and the KFI provides the specific amount, its 2015 equivalent, and the source. The <|retrieval|> token is inserted to indicate where additional information about the foundation might be needed.",
"Ideal Knowledge": "Information about the Frederik and Lena Meijer Foundation, including its history of supporting local infrastructure projects and typical donation amounts.",
"Key Factual Information": "$ 350,000 in 2015 ) from the Frederik and Lena Meijer Foundation",
"Prefix reformulated": "What was the value of the Frederik and Lena Meijer Foundation's contribution to the M-6 Trail project in 2015 dollars?",
"Prefix-intent": "and $ 300,000 ( equivalent to",
"Verification": "The KFI directly provides the equivalent value of the foundation's contribution in 2015 dollars, as prompted by the PI."
}
]
} |
In 2009 , the asphalt section of M @-@ 6 had to be repaired . This section of roadway between East Paris Avenue and 48th Street was rated poorly by the Michigan Infrastructure and Transportation Association , while the concrete west of Broadmoor Avenue had favorable marks . MDOT budgeted $ 2 million in repairs on top of previous crack @-@ related fixes that were handled by the original pavement contractor under a warranty in 2006 . The local press described the 4 @.@ 7 @-@ mile ( 7 @.@ 6 km ) stretch of road as " troublesome " in relation to pavement quality issues .
| [
{
"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 2009 , the asphalt section of M @-@ 6 had to be repaired . This section of roadway between East Paris Avenue and 48th Street was rated poorly by the Michigan Infrastructure and Transportation Association , while the concrete west of Broadmoor Avenue had favorable marks . MDOT budgeted $ 2 million in repairs on top of previous crack @-@ related fixes that were handled by the original pavement contractor under a warranty in 2006 . The local press described the 4 @.@ 7 @-@ mile ( 7 @.@ 6 km ) stretch of road as \" troublesome \" in relation to pavement quality issues . \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
} |
Legislation was signed by Governor Rick Snyder on December 27 , 2014 , to name the section of M @-@ 6 between Byron Center and Kalamazoo avenues the David John Warsen Memorial Highway . Warsen , a US Navy SEAL , was killed in a helicopter accident in Afghanistan in 2012 . This section of the highway was dedicated on August 15 , 2015 .
| [
{
"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 Legislation was signed by Governor Rick Snyder on December 27 , 2014 , to name the section of M @-@ 6 between Byron Center and Kalamazoo avenues the David John Warsen Memorial Highway . Warsen , a US Navy SEAL , was killed in a helicopter accident in Afghanistan in 2012 . This section of the highway was dedicated on August 15 , 2015 . \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
} |
" Hi , Infidelity " is the sixth episode of the third season of the American mystery television series Veronica Mars , and the fiftieth episode overall . Written by John Enbom and directed by Michael Fields , the episode premiered on The CW on November 17 , 2006 .
| [
{
"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 \" Hi , Infidelity \" is the sixth episode of the third season of the American mystery television series Veronica Mars , and the fiftieth episode overall . Written by John Enbom and directed by Michael Fields , the episode premiered on The CW on November 17 , 2006 . \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 series depicts the adventures of Veronica Mars ( Kristen Bell ) as she deals with life as a college student while moonlighting as a private detective . In this episode , Veronica investigates an accusation that she plagiarized a criminology paper only to find out that her teacher is having an affair with Mindy O 'Dell ( Jamie Ray Newman ) . Meanwhile , Stosh " Piz " Piznarski ( Chris Lowell ) invites Veronica to go bowling , and Wallace Fennel ( Percy Daggs III ) must choose between basketball and his studies .
| [
{
"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 series depicts the adventures of Veronica Mars ( Kristen Bell ) as she deals with life as a college student while moonlighting as a private detective . In this episode , Veronica investigates an accusation that she plagiarized a criminology paper only to find out that her teacher is having an affair with Mindy O 'Dell ( Jamie Ray Newman ) . Meanwhile , Stosh \" Piz \" Piznarski ( Chris Lowell ) invites Veronica to go bowling , and Wallace Fennel ( Percy Daggs III ) must choose between basketball and his studies . \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 episode features the return of Laura San Giacomo as Harmony Chase and the introduction of the recurring character of Max ( Adam Rose ) . San Giacomo and Colantoni , who had previously co @-@ starred on the sitcom Just Shoot Me ! , lobbied for more storylines together after rekindling their friendship during production of " Charlie Don 't Surf " , her first appearance . Rose 's character had been planned to begin a later romantic relationship with Mac since the writing of this episode and went on to fulfill this role in several future episodes . At the time of its initial broadcast , the episode was watched by 2 @.@ 75 million people and received mixed to positive reviews from television critics .
| [
{
"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 episode features the return of Laura San Giacomo as Harmony Chase and the introduction of the recurring character of Max ( Adam Rose ) . San Giacomo and Colantoni , who had previously co @-@ starred on the sitcom Just Shoot Me ! , lobbied for more storylines together after rekindling their friendship during production of \" Charlie Don 't Surf \" , her first appearance . Rose 's character had been planned to begin a later romantic relationship with Mac since the writing of this episode and went on to fulfill this role in several future episodes . At the time of its initial broadcast , the episode was watched by 2 @.@ 75 million people and received mixed to positive reviews from television critics . \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
} |
Following the events of the previous episode , Veronica confronts Claire Nordhouse ( Krista Kalmus ) for faking her rape after she publishes a story on it . Wallace Fennel and a lawyer are taken in to Dean O ’ Dell ’ s ( Ed Begley , Jr . ) office , where he is caught for cheating on a test . Veronica ’ s criminology paper is praised by her teacher , Hank ( Patrick Fabian ) leading her to praise him incessantly . Harmony Chase ( Laura San Giacomo ) , a former client who asked Keith Mars ( Enrico Colantoni ) to investigate her possibly adulterous husband , calls him and asks him on a date . Tim Foyle ( James Jordan ) , Hank 's teaching assistant , tells Veronica that she plagiarized her paper , even though she did not . Hank gives her three days to prove that she is innocent . Veronica learns that a student named Jeff Ratner accused her . Veronica goes to a computer student and ascertains the email address of the person who faked her paper . Stosh " Piz " Piznarski invites Veronica bowling , and she invites Parker ( Julie Gonzalo ) as well .
| [
{
"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 Following the events of the previous episode , Veronica confronts Claire Nordhouse ( Krista Kalmus ) for faking her rape after she publishes a story on it . Wallace Fennel and a lawyer are taken in to Dean O ’ Dell ’ s ( Ed Begley , Jr . ) office , where he is caught for cheating on a test . Veronica ’ s criminology paper is praised by her teacher , Hank ( Patrick Fabian ) leading her to praise him incessantly . Harmony Chase ( Laura San Giacomo ) , a former client who asked Keith Mars ( Enrico Colantoni ) to investigate her possibly adulterous husband , calls him and asks him on a date . Tim Foyle ( James Jordan ) , Hank 's teaching assistant , tells Veronica that she plagiarized her paper , even though she did not . Hank gives her three days to prove that she is innocent . Veronica learns that a student named Jeff Ratner accused her . Veronica goes to a computer student and ascertains the email address of the person who faked her paper . Stosh \" Piz \" Piznarski invites Veronica bowling , and she invites Parker ( Julie Gonzalo ) as well . \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
} |
Veronica gets caught snooping in the Dean ’ s office , although she makes up a hasty lie that she was searching for class notes , and it diverts his attention . Wallace decides not to drop the class on whose test he cheated . Veronica proves that the essay hers was supposedly plagiarized from was posted after she turned in her paper ; nevertheless , she still wants to find who framed her . Parker Lee , Veronica , Piz , and Logan ( Jason Dohring ) bowl and have fun together . When Veronica and Logan get room service , Veronica spots Jeff Ratner and questions him . Parker is romantically interested in Piz , and she tasks Veronica with talking to him about her . After talking to Mercer Hayes ( Ryan Devlin ) , Parker informs Veronica that she remembers Mercer 's cologne from the night of her rape . Veronica goes to Sheriff Lamb ( Michael Muhney ) with this news ; Wallace drops basketball to study for the class . Veronica runs into Keith at the hotel before learning that the room that belongs to " Rory Finch " is actually Hank , who is having an affair with Mrs. O ’ Dell .
| [
{
"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 Veronica gets caught snooping in the Dean ’ s office , although she makes up a hasty lie that she was searching for class notes , and it diverts his attention . Wallace decides not to drop the class on whose test he cheated . Veronica proves that the essay hers was supposedly plagiarized from was posted after she turned in her paper ; nevertheless , she still wants to find who framed her . Parker Lee , Veronica , Piz , and Logan ( Jason Dohring ) bowl and have fun together . When Veronica and Logan get room service , Veronica spots Jeff Ratner and questions him . Parker is romantically interested in Piz , and she tasks Veronica with talking to him about her . After talking to Mercer Hayes ( Ryan Devlin ) , Parker informs Veronica that she remembers Mercer 's cologne from the night of her rape . Veronica goes to Sheriff Lamb ( Michael Muhney ) with this news ; Wallace drops basketball to study for the class . Veronica runs into Keith at the hotel before learning that the room that belongs to \" Rory Finch \" is actually Hank , who is having an affair with Mrs. O ’ Dell . \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 their second date , Harmony suggests that they have sex , but Keith declines the proposal . On the way back , Keith ’ s becomes involved in a traffic collision , but the airbag catches him . This traumatic experience makes him go back to Harmony , and they sleep together . Veronica talks to Tim Foyle , who made her follow the trail of " Rory Finch " deliberately in order for her to discover the professor ’ s affair . Tim was the professor ’ s protégé , and he wants Veronica to discover Hank 's flaws before she becomes more involved with him . Logan runs up to Veronica and tells her that Mercer has been arrested for the rapes on campus , despite the fact that he believes that Mercer is innocent . Logan begs Veronica to defend Mercer , as he was with him the night of one of the rapes . However , he refuses to tell Veronica what they were doing .
| [
{
"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 their second date , Harmony suggests that they have sex , but Keith declines the proposal . On the way back , Keith ’ s becomes involved in a traffic collision , but the airbag catches him . This traumatic experience makes him go back to Harmony , and they sleep together . Veronica talks to Tim Foyle , who made her follow the trail of \" Rory Finch \" deliberately in order for her to discover the professor ’ s affair . Tim was the professor ’ s protégé , and he wants Veronica to discover Hank 's flaws before she becomes more involved with him . Logan runs up to Veronica and tells her that Mercer has been arrested for the rapes on campus , despite the fact that he believes that Mercer is innocent . Logan begs Veronica to defend Mercer , as he was with him the night of one of the rapes . However , he refuses to tell Veronica what they were doing . \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
} |
" Hi , Infidelity " was written by John Enbom and directed by Michael Fields , marking Enbom 's twelfth writing credit and Fields 's sixth directing credit for the series . The episode features the second of three appearances by Laura San Giacomo as Harmony Chase . She and Enrico Colantoni had become friends when they co @-@ starred on the sitcom Just Shoot Me ! . They rekindled their friendship during production of " Charlie Don 't Surf " , San Giacomo 's first appearance , so they lobbied for more storylines together . Thomas was open to the possibility , saying , " I wouldn 't hesitate to go back to this , because I was really happy with how it played out . "
| [
{
"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 \" Hi , Infidelity \" was written by John Enbom and directed by Michael Fields , marking Enbom 's twelfth writing credit and Fields 's sixth directing credit for the series . The episode features the second of three appearances by Laura San Giacomo as Harmony Chase . She and Enrico Colantoni had become friends when they co @-@ starred on the sitcom Just Shoot Me ! . They rekindled their friendship during production of \" Charlie Don 't Surf \" , San Giacomo 's first appearance , so they lobbied for more storylines together . Thomas was open to the possibility , saying , \" I wouldn 't hesitate to go back to this , because I was really happy with how it played out . \" \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
} |