text
stringlengths 101
3.69k
| messages
listlengths 1
1
| outputs
dict |
|---|---|---|
Observing their burdensome prosodic effect , the poet Matthew Arnold remarked , " these three lines by themselves take up nearly as much time as a whole book of the Iliad . " Many of the poem 's clauses carry over into the following line ; these enjambments emphasize Ulysses ' restlessness and dissatisfaction .
| [
{
"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 Observing their burdensome prosodic effect , the poet Matthew Arnold remarked , \" these three lines by themselves take up nearly as much time as a whole book of the Iliad . \" Many of the poem 's clauses carry over into the following line ; these enjambments emphasize Ulysses ' restlessness and dissatisfaction . \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|>Observing their burdensome prosodic effect , the poet Matthew Arnold remarked ,<|end-PI|> <|retrieval|> <|start-KFI|>\" these three lines by themselves take up nearly as much time as a whole book of the Iliad . \"<|end-KFI|> <|start-PI|>Many of the poem 's clauses carry over into the following line ; these enjambments emphasize<|end-PI|> <|retrieval|> <|start-KFI|>Ulysses ' restlessness and dissatisfaction<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces Matthew Arnold's observation about a prosodic effect, while the KFI provides his exact quote. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "A collection of Matthew Arnold's literary criticisms, specifically his comments on the prosody of epic poetry and comparisons to Homer's Iliad.",
"Key Factual Information": "\" these three lines by themselves take up nearly as much time as a whole book of the Iliad . \"",
"Prefix reformulated": "What did the poet Matthew Arnold remark about the burdensome prosodic effect?",
"Prefix-intent": "Observing their burdensome prosodic effect , the poet Matthew Arnold remarked ,",
"Verification": "The KFI directly provides Matthew Arnold's remark about the prosodic effect, as prompted by the PI."
},
{
"Explanation": "The PI describes the use of enjambments in the poem, while the KFI provides the specific emotional effects these enjambments create. The <|retrieval|> token is inserted to indicate where additional information about poetic techniques and their effects might be needed.",
"Ideal Knowledge": "An analysis of the poetic techniques used in the poem 'Ulysses', focusing on the use of enjambment and its effects on the portrayal of the main character's emotional state.",
"Key Factual Information": "Ulysses ' restlessness and dissatisfaction",
"Prefix reformulated": "What do the enjambments in the poem emphasize?",
"Prefix-intent": "Many of the poem 's clauses carry over into the following line ; these enjambments emphasize",
"Verification": "The KFI directly states what the enjambments emphasize, as prompted by the PI."
}
]
} |
The poem 's seventy lines of blank verse are presented as a dramatic monologue . Scholars disagree on how Ulysses ' speech functions in this format ; it is not necessarily clear to whom Ulysses is speaking , if anyone , and from what location . Some see the verse turning from a soliloquy to a public address , as Ulysses seems to speak to himself in the first movement , then to turn to an audience as he introduces his son , and then to relocate to the seashore where he addresses his mariners . In this interpretation , the comparatively direct and honest language of the first movement is set against the more politically minded tone of the last two movements . For example , the second paragraph ( 33 – 43 ) about Telemachus , in which Ulysses muses again about domestic life , is a " revised version [ of lines 1 – 5 ] for public consumption " : a " savage race " is revised to a " rugged people " .
| [
{
"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 poem 's seventy lines of blank verse are presented as a dramatic monologue . Scholars disagree on how Ulysses ' speech functions in this format ; it is not necessarily clear to whom Ulysses is speaking , if anyone , and from what location . Some see the verse turning from a soliloquy to a public address , as Ulysses seems to speak to himself in the first movement , then to turn to an audience as he introduces his son , and then to relocate to the seashore where he addresses his mariners . In this interpretation , the comparatively direct and honest language of the first movement is set against the more politically minded tone of the last two movements . For example , the second paragraph ( 33 – 43 ) about Telemachus , in which Ulysses muses again about domestic life , is a \" revised version [ of lines 1 – 5 ] for public consumption \" : a \" savage race \" is revised to a \" rugged people \" . \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 poem 's seventy lines of blank verse are <|start-PI|> presented as <|end-PI|> <|retrieval|> <|start-KFI|> a dramatic monologue <|end-KFI|> . <|start-PI|> Scholars disagree on how Ulysses ' speech functions in this format ; it is not necessarily clear <|end-PI|> <|retrieval|> <|start-KFI|> to whom Ulysses is speaking , if anyone , and from what location <|end-KFI|> . <|start-PI|> Some see the verse turning from a soliloquy to a public address , as Ulysses seems <|end-PI|> <|retrieval|> <|start-KFI|> to speak to himself in the first movement , then to turn to an audience as he introduces his son , and then to relocate to the seashore where he addresses his mariners <|end-KFI|> . In this interpretation , the comparatively direct and honest language of the first movement is set against the more politically minded tone of the last two movements . <|start-PI|> For example , the second paragraph ( 33 – 43 ) about Telemachus , in which Ulysses muses again about domestic life , is <|end-PI|> <|retrieval|> <|start-KFI|> a \" revised version [ of lines 1 – 5 ] for public consumption \" : a \" savage race \" is revised to a \" rugged people \" <|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the poem's structure and sets up a query about its presentation, while the KFI provides the specific format. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "Information about poetic forms, specifically focusing on dramatic monologues and their characteristics in relation to blank verse poems.",
"Key Factual Information": "a dramatic monologue",
"Prefix reformulated": "How are the seventy lines of blank verse in the poem presented?",
"Prefix-intent": "The poem 's seventy lines of blank verse are presented as",
"Verification": "The KFI directly provides the presentation format of the poem as requested in the PI."
},
{
"Explanation": "The PI sets up the scholarly disagreement, and the KFI provides specific points of contention. The <|retrieval|> token is inserted to indicate where additional information about scholarly interpretations might be needed.",
"Ideal Knowledge": "Scholarly debates and interpretations of Ulysses' speech in the poem, focusing on the ambiguity of the intended audience and the speaker's location.",
"Key Factual Information": "to whom Ulysses is speaking , if anyone , and from what location",
"Prefix reformulated": "What aspects of Ulysses' speech in the poem are unclear to scholars?",
"Prefix-intent": "Scholars disagree on how Ulysses ' speech functions in this format ; it is not necessarily clear",
"Verification": "The KFI directly addresses the unclear aspects of Ulysses' speech as prompted by the PI."
},
{
"Explanation": "The PI introduces an interpretation of the verse's structure, and the KFI provides specific details about the perceived changes. The <|retrieval|> token is inserted to indicate where additional information about the poem's structure might be needed.",
"Ideal Knowledge": "Detailed analysis of the structure and progression of Ulysses' speech in the poem, highlighting the shifts in audience and setting throughout the monologue.",
"Key Factual Information": "to speak to himself in the first movement , then to turn to an audience as he introduces his son , and then to relocate to the seashore where he addresses his mariners",
"Prefix reformulated": "How does Ulysses' speech in the poem change according to some interpretations?",
"Prefix-intent": "Some see the verse turning from a soliloquy to a public address , as Ulysses seems",
"Verification": "The KFI directly describes the perceived changes in Ulysses' speech as prompted by the PI."
},
{
"Explanation": "The PI introduces a specific part of the poem, and the KFI provides details about how it's interpreted as a revision. The <|retrieval|> token is inserted to indicate where additional information about textual analysis might be needed.",
"Ideal Knowledge": "Textual analysis of the specific lines mentioned in the poem, comparing the language used in different sections and discussing the implications of these revisions.",
"Key Factual Information": "a \" revised version [ of lines 1 – 5 ] for public consumption \" : a \" savage race \" is revised to a \" rugged people \"",
"Prefix reformulated": "How is the second paragraph about Telemachus in Ulysses' speech characterized?",
"Prefix-intent": "For example , the second paragraph ( 33 – 43 ) about Telemachus , in which Ulysses muses again about domestic life , is",
"Verification": "The KFI directly provides the characterization of the second paragraph as requested in the PI."
}
]
} |
The ironic interpretations of " Ulysses " may be the result of the modern tendency to consider the narrator of a dramatic monologue as necessarily " unreliable " . According to critic Dwight Culler , the poem has been a victim of revisionist readings in which the reader expects to reconstruct the truth from a misleading narrator 's accidental revelations . ( Compare the more obvious use of this approach in Robert Browning 's " My Last Duchess " . ) Culler himself views " Ulysses " as a dialectic in which the speaker weighs the virtues of a contemplative and an active approach to life ; Ulysses moves through four emotional stages that are self @-@ revelatory , not ironic : beginning with his rejection of the barren life to which he has returned in Ithaca , he then fondly recalls his heroic past , recognizes the validity of Telemachus ' method of governing , and with these thoughts plans another journey .
| [
{
"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 ironic interpretations of \" Ulysses \" may be the result of the modern tendency to consider the narrator of a dramatic monologue as necessarily \" unreliable \" . According to critic Dwight Culler , the poem has been a victim of revisionist readings in which the reader expects to reconstruct the truth from a misleading narrator 's accidental revelations . ( Compare the more obvious use of this approach in Robert Browning 's \" My Last Duchess \" . ) Culler himself views \" Ulysses \" as a dialectic in which the speaker weighs the virtues of a contemplative and an active approach to life ; Ulysses moves through four emotional stages that are self @-@ revelatory , not ironic : beginning with his rejection of the barren life to which he has returned in Ithaca , he then fondly recalls his heroic past , recognizes the validity of Telemachus ' method of governing , and with these thoughts plans another journey . \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 ironic interpretations of \" Ulysses \" may be the result of <|start-PI|> the modern tendency to consider the narrator of a dramatic monologue as necessarily <|end-PI|> <|retrieval|> <|start-KFI|> \" unreliable \" <|end-KFI|> . <|start-PI|> According to critic Dwight Culler , the poem has been a victim of revisionist readings in which <|end-PI|> <|retrieval|> <|start-KFI|> the reader expects to reconstruct the truth from a misleading narrator 's accidental revelations <|end-KFI|> . ( Compare the more obvious use of this approach in Robert Browning 's \" My Last Duchess \" . ) <|start-PI|> Culler himself views \" Ulysses \" as <|end-PI|> <|retrieval|> <|start-KFI|> a dialectic in which the speaker weighs the virtues of a contemplative and an active approach to life <|end-KFI|> ; <|start-PI|> Ulysses moves through four emotional stages that are <|end-PI|> <|retrieval|> <|start-KFI|> self @-@ revelatory , not ironic <|end-KFI|> : beginning with his rejection of the barren life to which he has returned in Ithaca , he then fondly recalls his heroic past , recognizes the validity of Telemachus ' method of governing , and with these thoughts plans another journey .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the topic of ironic interpretations of 'Ulysses', while the KFI provides the specific reason for these interpretations. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the statement.",
"Ideal Knowledge": "Information on literary criticism trends, particularly regarding the interpretation of dramatic monologues and the concept of unreliable narrators in modern literary analysis.",
"Key Factual Information": "the modern tendency to consider the narrator of a dramatic monologue as necessarily \" unreliable \"",
"Prefix reformulated": "What may be the reason for the ironic interpretations of \"Ulysses\"?",
"Prefix-intent": "The ironic interpretations of \" Ulysses \" may be the result of",
"Verification": "The KFI directly provides the reason for the ironic interpretations mentioned in the PI."
},
{
"Explanation": "The PI introduces Culler's perspective on revisionist readings of the poem, while the KFI provides the specific nature of these readings. The <|retrieval|> token is inserted to indicate where additional information about Culler's analysis might be needed.",
"Ideal Knowledge": "Detailed explanation of Dwight Culler's literary criticism, focusing on his analysis of 'Ulysses' and the concept of revisionist readings in poetry interpretation.",
"Key Factual Information": "the reader expects to reconstruct the truth from a misleading narrator 's accidental revelations",
"Prefix reformulated": "What kind of revisionist readings has \"Ulysses\" been subjected to, according to Dwight Culler?",
"Prefix-intent": "According to critic Dwight Culler , the poem has been a victim of revisionist readings in which",
"Verification": "The KFI directly describes the type of revisionist readings mentioned by Culler in the PI."
},
{
"Explanation": "The PI sets up Culler's view of the poem, while the KFI provides his specific interpretation. The <|retrieval|> token is inserted to indicate where additional information about Culler's analysis might be needed.",
"Ideal Knowledge": "Detailed explanation of Culler's interpretation of 'Ulysses', including his analysis of the poem's themes and structure, particularly the dialectic between contemplative and active approaches to life.",
"Key Factual Information": "a dialectic in which the speaker weighs the virtues of a contemplative and an active approach to life",
"Prefix reformulated": "How does Culler view \"Ulysses\"?",
"Prefix-intent": "Culler himself views \" Ulysses \" as",
"Verification": "The KFI directly provides Culler's interpretation of 'Ulysses' as mentioned in the PI."
},
{
"Explanation": "The PI introduces the concept of Ulysses' emotional stages, while the KFI provides their key characteristic. The <|retrieval|> token is inserted to indicate where additional information about these stages might be needed.",
"Ideal Knowledge": "Detailed analysis of the character development in 'Ulysses', focusing on the four emotional stages and their self-revelatory nature, contrasting this with ironic interpretations.",
"Key Factual Information": "self @-@ revelatory , not ironic",
"Prefix reformulated": "What are the characteristics of the emotional stages Ulysses goes through?",
"Prefix-intent": "Ulysses moves through four emotional stages that are",
"Verification": "The KFI directly describes the nature of the emotional stages mentioned in the PI."
}
]
} |
Tennyson completed the poem on 20 October 1833 , but it was not published until 1842 , in his second collection of Poems . Unlike many of Tennyson 's other important poems , " Ulysses " was not revised after its publication .
| [
{
"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 Tennyson completed the poem on 20 October 1833 , but it was not published until 1842 , in his second collection of Poems . Unlike many of Tennyson 's other important poems , \" Ulysses \" was not revised after its publication . \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
} |
Tennyson originally blocked out the poem in four paragraphs , broken before lines 6 , 33 and 44 . In this structure , the first and third paragraphs are thematically parallel , but may be read as interior and exterior monologues , respectively . However , the poem is often printed with the first paragraph break omitted .
| [
{
"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 Tennyson originally blocked out the poem in four paragraphs , broken before lines 6 , 33 and 44 . In this structure , the first and third paragraphs are thematically parallel , but may be read as interior and exterior monologues , respectively . However , the poem is often printed with the first paragraph break omitted . \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|>Tennyson originally blocked out the poem in<|end-PI|> <|retrieval|> <|start-KFI|>four paragraphs , broken before lines 6 , 33 and 44<|end-KFI|> . <|start-PI|>In this structure , the first and third paragraphs are<|end-PI|> <|retrieval|> <|start-KFI|>thematically parallel , but may be read as interior and exterior monologues , respectively<|end-KFI|> . <|start-PI|>However , the poem is often printed with<|end-PI|> <|retrieval|> <|start-KFI|>the first paragraph break omitted<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces Tennyson's original structuring of the poem, while the KFI provides the specific paragraph count and line numbers. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "Detailed information about Tennyson's original manuscript or drafts of the poem, including the specific line numbers where paragraph breaks were placed.",
"Key Factual Information": "four paragraphs , broken before lines 6 , 33 and 44",
"Prefix reformulated": "How did Tennyson originally structure the poem?",
"Prefix-intent": "Tennyson originally blocked out the poem in",
"Verification": "The KFI directly provides the structure information requested in the PI."
},
{
"Explanation": "The PI sets up a query about the relationship between specific paragraphs, and the KFI provides information about their thematic connection and possible interpretation. The <|retrieval|> token is inserted to indicate where additional literary analysis might be needed.",
"Ideal Knowledge": "Literary analysis of Tennyson's poem, focusing on the thematic parallels between paragraphs and the concept of interior and exterior monologues in poetry.",
"Key Factual Information": "thematically parallel , but may be read as interior and exterior monologues , respectively",
"Prefix reformulated": "What is the relationship between the first and third paragraphs in Tennyson's original structure?",
"Prefix-intent": "In this structure , the first and third paragraphs are",
"Verification": "The KFI directly describes the relationship between the paragraphs as prompted by the PI."
},
{
"Explanation": "The PI introduces a difference in how the poem is often printed, and the KFI specifies the exact change. The <|retrieval|> token is inserted to indicate where additional information about published versions might be needed.",
"Ideal Knowledge": "Information on various published versions of Tennyson's poem, detailing differences in paragraph structure compared to the original manuscript.",
"Key Factual Information": "the first paragraph break omitted",
"Prefix reformulated": "How is the poem often printed differently from Tennyson's original structure?",
"Prefix-intent": "However , the poem is often printed with",
"Verification": "The KFI directly answers how the printed version differs from the original structure, as prompted by the PI."
}
]
} |
Tennyson penned " Ulysses " after the death of his close Cambridge friend , the poet Arthur Henry Hallam ( 1811 – 1833 ) , with whom Tennyson had a strong emotional bond . The two friends had spent much time discussing poetry and philosophy , writing verse , and travelling in southern France , the Pyrenees , and Germany . Tennyson considered Hallam destined for greatness , perhaps as a statesman .
| [
{
"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 Tennyson penned \" Ulysses \" after the death of his close Cambridge friend , the poet Arthur Henry Hallam ( 1811 – 1833 ) , with whom Tennyson had a strong emotional bond . The two friends had spent much time discussing poetry and philosophy , writing verse , and travelling in southern France , the Pyrenees , and Germany . Tennyson considered Hallam destined for greatness , perhaps as a statesman . \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
} |
When Tennyson heard on 1 October 1833 of his friend 's death , he was living in Somersby , Lincolnshire , in cramped quarters with his mother and nine of his ten siblings . His father had died in 1831 , requiring Tennyson to return home and take responsibility for the family . Tennyson 's friends were becoming increasingly concerned about his mental and physical health during this time . The family had little income , and three of Tennyson 's brothers were mentally ill . Just as Tennyson 's outlook was improving — he was adjusting to his new domestic duties , regaining contact with friends , and had published his 1832 book of poems — the news of Hallam 's death arrived . Tennyson shared his grief with his sister , Emily , who had been engaged to Hallam .
| [
{
"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 When Tennyson heard on 1 October 1833 of his friend 's death , he was living in Somersby , Lincolnshire , in cramped quarters with his mother and nine of his ten siblings . His father had died in 1831 , requiring Tennyson to return home and take responsibility for the family . Tennyson 's friends were becoming increasingly concerned about his mental and physical health during this time . The family had little income , and three of Tennyson 's brothers were mentally ill . Just as Tennyson 's outlook was improving — he was adjusting to his new domestic duties , regaining contact with friends , and had published his 1832 book of poems — the news of Hallam 's death arrived . Tennyson shared his grief with his sister , Emily , who had been engaged to Hallam . \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|>When Tennyson heard on 1 October 1833 of his friend's death, he was living in<|end-PI|> <|retrieval|> <|start-KFI|>Somersby, Lincolnshire, in cramped quarters with his mother and nine of his ten siblings<|end-KFI|>. <|start-PI|>His father had died in 1831, requiring Tennyson to<|end-PI|> <|retrieval|> <|start-KFI|>return home and take responsibility for the family<|end-KFI|>. Tennyson's friends were becoming increasingly concerned about his mental and physical health during this time. <|start-PI|>The family had little income, and three of Tennyson's brothers were<|end-PI|> <|retrieval|> <|start-KFI|>mentally ill<|end-KFI|>. <|start-PI|>Just as Tennyson's outlook was improving — he was adjusting to his new domestic duties, regaining contact with friends, and had published his<|end-PI|> <|retrieval|> <|start-KFI|>1832 book of poems<|end-KFI|> — the news of Hallam's death arrived. <|start-PI|>Tennyson shared his grief with his sister, Emily, who had been<|end-PI|> <|retrieval|> <|start-KFI|>engaged to Hallam<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of Tennyson receiving news of his friend's death and asks about his location, while the KFI provides specific details about where and with whom he was living. The <|retrieval|> token is inserted to indicate where additional information about Tennyson's living situation might be needed.",
"Ideal Knowledge": "Detailed information about Tennyson's residence in 1833, including the location, family members present, and living conditions in Somersby, Lincolnshire.",
"Key Factual Information": "Somersby, Lincolnshire, in cramped quarters with his mother and nine of his ten siblings",
"Prefix reformulated": "Where was Tennyson living when he heard of his friend's death on 1 October 1833?",
"Prefix-intent": "When Tennyson heard on 1 October 1833 of his friend's death, he was living in",
"Verification": "The KFI directly provides the location and living situation that the PI inquires about."
},
{
"Explanation": "The PI introduces the event of Tennyson's father's death and sets up a query about its consequences, while the KFI provides the specific actions Tennyson had to take. The <|retrieval|> token is inserted to indicate where additional information about Tennyson's family responsibilities might be needed.",
"Ideal Knowledge": "Information about the impact of Tennyson's father's death on the family, including Tennyson's responsibilities and the changes in his life circumstances.",
"Key Factual Information": "return home and take responsibility for the family",
"Prefix reformulated": "What did Tennyson have to do after his father died in 1831?",
"Prefix-intent": "His father had died in 1831, requiring Tennyson to",
"Verification": "The KFI directly states the actions Tennyson had to take following his father's death, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of the family's financial situation and introduces a query about Tennyson's brothers, while the KFI provides the specific information about their mental health. The <|retrieval|> token is inserted to indicate where additional information about the brothers' conditions might be needed.",
"Ideal Knowledge": "Detailed information about the mental health issues in the Tennyson family, specifically focusing on the three brothers mentioned and how their condition affected the family dynamics.",
"Key Factual Information": "mentally ill",
"Prefix reformulated": "What was the condition of three of Tennyson's brothers?",
"Prefix-intent": "The family had little income, and three of Tennyson's brothers were",
"Verification": "The KFI directly states the condition of Tennyson's brothers, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of Tennyson's improving situation and leads to a query about his publication, while the KFI provides the specific information about the book of poems. The <|retrieval|> token is inserted to indicate where additional information about this publication might be needed.",
"Ideal Knowledge": "Details about Tennyson's 1832 book of poems, including its title, content, reception, and significance in his career.",
"Key Factual Information": "1832 book of poems",
"Prefix reformulated": "What had Tennyson published as his outlook was improving?",
"Prefix-intent": "Just as Tennyson's outlook was improving — he was adjusting to his new domestic duties, regaining contact with friends, and had published his",
"Verification": "The KFI directly provides information about Tennyson's publication, as prompted by the PI."
},
{
"Explanation": "The PI introduces Tennyson sharing his grief with his sister and sets up a query about her connection to the deceased friend, while the KFI provides the specific information about her engagement to Hallam. The <|retrieval|> token is inserted to indicate where additional information about this relationship might be needed.",
"Ideal Knowledge": "Information about the engagement between Emily Tennyson and Arthur Hallam, including details about their relationship and how it affected the Tennyson family.",
"Key Factual Information": "engaged to Hallam",
"Prefix reformulated": "What was the relationship between Tennyson's sister Emily and his deceased friend?",
"Prefix-intent": "Tennyson shared his grief with his sister, Emily, who had been",
"Verification": "The KFI directly states the relationship between Emily and Hallam, as prompted by the PI."
}
]
} |
According to Victorian scholar Linda Hughes , the emotional gulf between the state of his domestic affairs and the loss of his special friendship informs the reading of " Ulysses " — particularly its treatment of domesticity . At one moment , Ulysses ' discontent seems to mirror that of Tennyson , who would have been frustrated with managing the house in such a state of grief . At the next , Ulysses is determined to transcend his age and his environment by travelling again . It may be that Ulysses ' determination to defy circumstance attracted Tennyson to the myth ; he said that the poem " gave my feeling about the need of going forward and braving the struggle of life " . On another occasion , the poet stated , " There is more about myself in Ulysses , which was written under the sense of loss and that all had gone by , but that still life must be fought out to the end . It was more written with the feeling of his loss upon me than many poems in In Memoriam . " Hallam 's death influenced much of Tennyson 's poetry , including perhaps his most highly regarded work , In Memoriam A.H.H. , begun in 1833 and completed seventeen years later .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n According to Victorian scholar Linda Hughes , the emotional gulf between the state of his domestic affairs and the loss of his special friendship informs the reading of \" Ulysses \" — particularly its treatment of domesticity . At one moment , Ulysses ' discontent seems to mirror that of Tennyson , who would have been frustrated with managing the house in such a state of grief . At the next , Ulysses is determined to transcend his age and his environment by travelling again . It may be that Ulysses ' determination to defy circumstance attracted Tennyson to the myth ; he said that the poem \" gave my feeling about the need of going forward and braving the struggle of life \" . On another occasion , the poet stated , \" There is more about myself in Ulysses , which was written under the sense of loss and that all had gone by , but that still life must be fought out to the end . It was more written with the feeling of his loss upon me than many poems in In Memoriam . \" Hallam 's death influenced much of Tennyson 's poetry , including perhaps his most highly regarded work , In Memoriam A.H.H. , begun in 1833 and completed seventeen years later . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "According to Victorian scholar Linda Hughes , <|start-PI|> the emotional gulf between the state of his domestic affairs and the loss of his special friendship informs the reading of <|end-PI|> <|retrieval|> <|start-KFI|> \" Ulysses \" — particularly its treatment of domesticity <|end-KFI|> . At one moment , Ulysses ' discontent seems to mirror that of Tennyson , who would have been frustrated with managing the house in such a state of grief . At the next , <|start-PI|> Ulysses is determined to <|end-PI|> <|retrieval|> <|start-KFI|> transcend his age and his environment by travelling again <|end-KFI|> . It may be that Ulysses ' determination to defy circumstance attracted <|start-PI|> Tennyson to the myth ; he said that the poem <|end-PI|> <|retrieval|> <|start-KFI|> \" gave my feeling about the need of going forward and braving the struggle of life \" <|end-KFI|> . <|start-PI|> On another occasion , the poet stated , <|end-PI|> <|retrieval|> <|start-KFI|> \" There is more about myself in Ulysses , which was written under the sense of loss and that all had gone by , but that still life must be fought out to the end . It was more written with the feeling of his loss upon me than many poems in In Memoriam . \" <|end-KFI|> <|start-PI|> Hallam 's death influenced much of Tennyson 's poetry , including perhaps his most highly regarded work , <|end-PI|> <|retrieval|> <|start-KFI|> In Memoriam A.H.H. , begun in 1833 and completed seventeen years later <|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces Linda Hughes' perspective on 'Ulysses', while the KFI specifies what aspect of the poem is informed by Tennyson's personal circumstances. The <|retrieval|> token is inserted to indicate where additional scholarly analysis might be needed.",
"Ideal Knowledge": "A detailed analysis of Linda Hughes' interpretation of Tennyson's 'Ulysses', focusing on the poem's treatment of domesticity and its connection to Tennyson's personal life.",
"Key Factual Information": "\" Ulysses \" — particularly its treatment of domesticity",
"Prefix reformulated": "What does Linda Hughes say informs the reading of Tennyson's 'Ulysses'?",
"Prefix-intent": "According to Victorian scholar Linda Hughes , the emotional gulf between the state of his domestic affairs and the loss of his special friendship informs the reading of",
"Verification": "The KFI directly provides the information about what informs the reading of 'Ulysses' as stated by Linda Hughes in the PI."
},
{
"Explanation": "The PI sets up the contrast between Tennyson's grief and Ulysses' determination, while the KFI specifies Ulysses' intended action. The <|retrieval|> token is inserted to indicate where additional information about the poem's themes might be needed.",
"Ideal Knowledge": "An analysis of the character Ulysses in Tennyson's poem, focusing on his desire to travel and escape his current situation, and how this reflects Tennyson's own feelings.",
"Key Factual Information": "transcend his age and his environment by travelling again",
"Prefix reformulated": "What is Ulysses determined to do in Tennyson's poem?",
"Prefix-intent": "Tennyson , who would have been frustrated with managing the house in such a state of grief . At the next , Ulysses is determined to",
"Verification": "The KFI directly provides Ulysses' determination as described in the poem, which is prompted by the PI."
},
{
"Explanation": "The PI introduces Tennyson's perspective on the poem, while the KFI provides his direct quote. The <|retrieval|> token is inserted to indicate where additional context about Tennyson's thoughts on the poem might be needed.",
"Ideal Knowledge": "A collection of Tennyson's personal statements and reflections on his poem 'Ulysses', particularly those relating to its themes of perseverance and life's struggles.",
"Key Factual Information": "\" gave my feeling about the need of going forward and braving the struggle of life \"",
"Prefix reformulated": "What did Tennyson say about the poem 'Ulysses'?",
"Prefix-intent": "Tennyson to the myth ; he said that the poem",
"Verification": "The KFI directly quotes Tennyson's statement about the poem, as prompted by the PI."
},
{
"Explanation": "The PI introduces another statement by Tennyson, while the KFI provides his direct quote relating 'Ulysses' to his personal experiences. The <|retrieval|> token is inserted to indicate where additional context about Tennyson's life and works might be needed.",
"Ideal Knowledge": "A comprehensive collection of Tennyson's statements about 'Ulysses', particularly those that relate the poem to his personal experiences of loss and his philosophy on facing life's challenges.",
"Key Factual Information": "\" There is more about myself in Ulysses , which was written under the sense of loss and that all had gone by , but that still life must be fought out to the end . It was more written with the feeling of his loss upon me than many poems in In Memoriam . \"",
"Prefix reformulated": "What did Tennyson state about 'Ulysses' on another occasion?",
"Prefix-intent": "On another occasion , the poet stated ,",
"Verification": "The KFI directly quotes Tennyson's statement about 'Ulysses', as prompted by the PI."
},
{
"Explanation": "The PI introduces the influence of Hallam's death on Tennyson's poetry, while the KFI specifies his most highly regarded work and its composition timeline. The <|retrieval|> token is inserted to indicate where additional information about 'In Memoriam A.H.H.' might be needed.",
"Ideal Knowledge": "Detailed information about Tennyson's 'In Memoriam A.H.H.', including its inspiration, composition period, and its significance in Tennyson's body of work.",
"Key Factual Information": "In Memoriam A.H.H. , begun in 1833 and completed seventeen years later",
"Prefix reformulated": "What is Tennyson's most highly regarded work influenced by Hallam's death?",
"Prefix-intent": "Hallam 's death influenced much of Tennyson 's poetry , including perhaps his most highly regarded work ,",
"Verification": "The KFI directly provides the title of Tennyson's highly regarded work and its timeframe, as prompted by the PI."
}
]
} |
Other critics find stylistic incongruities between the poem and its author that make " Ulysses " exceptional . W. W. Robson writes , " Tennyson , the responsible social being , the admirably serious and ' committed ' individual , is uttering strenuous sentiments in the accent of Tennyson the most un @-@ strenuous , lonely and poignant of poets . " He finds that Tennyson 's two widely noted personae , the " responsible social being " and the melancholic poet , meet uniquely in " Ulysses " , yet seem not to recognize each other within the text .
| [
{
"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 Other critics find stylistic incongruities between the poem and its author that make \" Ulysses \" exceptional . W. W. Robson writes , \" Tennyson , the responsible social being , the admirably serious and ' committed ' individual , is uttering strenuous sentiments in the accent of Tennyson the most un @-@ strenuous , lonely and poignant of poets . \" He finds that Tennyson 's two widely noted personae , the \" responsible social being \" and the melancholic poet , meet uniquely in \" Ulysses \" , yet seem not to recognize each other within the text . \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|>Other critics find stylistic incongruities between the poem and its author that make<|end-PI|> <|retrieval|> <|start-KFI|>\" Ulysses \" exceptional<|end-KFI|>. <|start-PI|>W. W. Robson writes , \" Tennyson , the responsible social being , the admirably serious and ' committed ' individual , is uttering strenuous sentiments in the accent of<|end-PI|> <|retrieval|> <|start-KFI|>Tennyson the most un @-@ strenuous , lonely and poignant of poets<|end-KFI|>. \" <|start-PI|>He finds that Tennyson 's two widely noted personae , the \" responsible social being \" and the melancholic poet , meet uniquely in<|end-PI|> <|retrieval|> <|start-KFI|>\" Ulysses \"<|end-KFI|> , yet seem not to recognize each other within the text.",
"none": null,
"results": [
{
"Explanation": "The PI introduces the idea of critics finding stylistic differences, and the KFI confirms that these differences make 'Ulysses' exceptional. The <|retrieval|> token is inserted to indicate where additional information about these stylistic incongruities might be needed.",
"Ideal Knowledge": "A detailed analysis of Tennyson's typical writing style compared to the style used in 'Ulysses', highlighting the specific differences that make the poem stand out from his other works.",
"Key Factual Information": "\" Ulysses \" exceptional",
"Prefix reformulated": "What stylistic incongruities do critics find between the poem 'Ulysses' and Tennyson that make the poem exceptional?",
"Prefix-intent": "Other critics find stylistic incongruities between the poem and its author that make",
"Verification": "The KFI directly states that 'Ulysses' is exceptional, which is the result of the stylistic incongruities mentioned in the PI."
},
{
"Explanation": "The PI sets up Robson's observation about Tennyson's typical voice, while the KFI provides the contrasting characteristics found in 'Ulysses'. The <|retrieval|> token is inserted to indicate where additional information about Tennyson's poetic styles might be needed.",
"Ideal Knowledge": "An analysis of Tennyson's poetic styles, comparing his socially responsible and committed voice with his more melancholic and introspective tone, specifically in the context of 'Ulysses'.",
"Key Factual Information": "Tennyson the most un @-@ strenuous , lonely and poignant of poets",
"Prefix reformulated": "How does W. W. Robson describe the contrast in Tennyson's voice in 'Ulysses'?",
"Prefix-intent": "W. W. Robson writes , \" Tennyson , the responsible social being , the admirably serious and ' committed ' individual , is uttering strenuous sentiments in the accent of",
"Verification": "The KFI directly contrasts with the description of Tennyson in the PI, providing the specific characteristics that create the incongruity."
},
{
"Explanation": "The PI introduces Robson's observation about Tennyson's two personae meeting, and the KFI specifies that this occurs in 'Ulysses'. The <|retrieval|> token is inserted to indicate where additional information about how these personae manifest in the poem might be needed.",
"Ideal Knowledge": "A detailed analysis of 'Ulysses', highlighting specific passages or elements that demonstrate the convergence of Tennyson's socially responsible and melancholic poetic voices.",
"Key Factual Information": "\" Ulysses \"",
"Prefix reformulated": "In which poem do Tennyson's two widely noted personae meet uniquely, according to Robson?",
"Prefix-intent": "He finds that Tennyson 's two widely noted personae , the \" responsible social being \" and the melancholic poet , meet uniquely in",
"Verification": "The KFI directly answers where the two personae meet, as prompted by the PI."
}
]
} |
Tennyson adopts aspects of the Ulysses character and narrative from many sources ; his treatment of Ulysses is the first modern account . The ancient Greek poet Homer introduced Ulysses ( Odysseus in Greek ) , and many later poets took up the character , including Euripides , Horace , Dante , William Shakespeare , and Alexander Pope . Homer 's Odyssey provides the poem 's narrative background : in its eleventh book the prophet Tiresias foretells that Ulysses will return to Ithaca after a difficult voyage , then begin a new , mysterious voyage , and later die a peaceful , " unwarlike " death that comes vaguely " from the sea " . At the conclusion of Tennyson 's poem , his Ulysses is contemplating undertaking this new voyage .
| [
{
"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 Tennyson adopts aspects of the Ulysses character and narrative from many sources ; his treatment of Ulysses is the first modern account . The ancient Greek poet Homer introduced Ulysses ( Odysseus in Greek ) , and many later poets took up the character , including Euripides , Horace , Dante , William Shakespeare , and Alexander Pope . Homer 's Odyssey provides the poem 's narrative background : in its eleventh book the prophet Tiresias foretells that Ulysses will return to Ithaca after a difficult voyage , then begin a new , mysterious voyage , and later die a peaceful , \" unwarlike \" death that comes vaguely \" from the sea \" . At the conclusion of Tennyson 's poem , his Ulysses is contemplating undertaking this new voyage . \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|>Tennyson adopts aspects of the Ulysses character and narrative from many sources ; his treatment of Ulysses is<|end-PI|> <|retrieval|> <|start-KFI|>the first modern account<|end-KFI|>. <|start-PI|>The ancient Greek poet Homer introduced Ulysses ( Odysseus in Greek ) , and many later poets took up the character , including<|end-PI|> <|retrieval|> <|start-KFI|>Euripides , Horace , Dante , William Shakespeare , and Alexander Pope<|end-KFI|>. <|start-PI|>Homer 's Odyssey provides the poem 's narrative background : in its eleventh book the prophet Tiresias foretells that Ulysses will<|end-PI|> <|retrieval|> <|start-KFI|>return to Ithaca after a difficult voyage , then begin a new , mysterious voyage , and later die a peaceful , \" unwarlike \" death that comes vaguely \" from the sea \"<|end-KFI|>. At the conclusion of Tennyson 's poem , his Ulysses is contemplating undertaking this new voyage.",
"none": null,
"results": [
{
"Explanation": "The PI introduces Tennyson's approach to Ulysses, and the KFI provides the significant fact about its modernity. The <|retrieval|> token is inserted to indicate where additional information might be needed to support this claim.",
"Ideal Knowledge": "A chronological list of literary works featuring Ulysses, highlighting Tennyson's version as the first modern interpretation and explaining how it differs from earlier accounts.",
"Key Factual Information": "the first modern account",
"Prefix reformulated": "What is Tennyson's treatment of Ulysses?",
"Prefix-intent": "Tennyson adopts aspects of the Ulysses character and narrative from many sources ; his treatment of Ulysses is",
"Verification": "The KFI directly provides information about Tennyson's treatment of Ulysses as requested in the PI."
},
{
"Explanation": "The PI sets up the context of Homer's introduction of Ulysses and mentions that other poets used the character, while the KFI provides specific names. The <|retrieval|> token is inserted to indicate where additional information about these poets and their works might be needed.",
"Ideal Knowledge": "A comprehensive list of poets and writers who have featured Ulysses in their works, including brief descriptions of how each author portrayed the character and how their interpretations differed or built upon previous versions.",
"Key Factual Information": "Euripides , Horace , Dante , William Shakespeare , and Alexander Pope",
"Prefix reformulated": "Which poets took up the character of Ulysses after Homer?",
"Prefix-intent": "The ancient Greek poet Homer introduced Ulysses ( Odysseus in Greek ) , and many later poets took up the character , including",
"Verification": "The KFI directly lists the poets who used the Ulysses character after Homer, as prompted by the PI."
},
{
"Explanation": "The PI introduces Tiresias' prophecy from Homer's Odyssey, and the KFI provides the specific details of the prophecy. The <|retrieval|> token is inserted to indicate where additional information about the prophecy and its significance might be needed.",
"Ideal Knowledge": "A detailed summary of Tiresias' prophecy in Homer's Odyssey, including its context, exact wording, and interpretations by scholars. This should also include information on how later authors, including Tennyson, incorporated or reinterpreted this prophecy in their works.",
"Key Factual Information": "return to Ithaca after a difficult voyage , then begin a new , mysterious voyage , and later die a peaceful , \" unwarlike \" death that comes vaguely \" from the sea \"",
"Prefix reformulated": "What does Tiresias foretell about Ulysses in Homer's Odyssey?",
"Prefix-intent": "Homer 's Odyssey provides the poem 's narrative background : in its eleventh book the prophet Tiresias foretells that Ulysses will",
"Verification": "The KFI directly provides the prophecy about Ulysses' future as foretold by Tiresias in the Odyssey, as prompted by the PI."
}
]
} |
Tennyson 's character , however , is not the lover of public affairs seen in Homer 's poems . Rather , " Ulisse " from Dante 's Inferno is Tennyson 's main source for the character , which has an important effect on the poem 's interpretation . Ulisse recalls his voyage in the Inferno 's 26th canto , in which he is condemned to the Eighth Circle of false counsellors for misusing his gift of reason . Dante treats Ulisse , with his " zeal … / T 'explore the world " , as an evil counsellor who lusts for adventure at the expense of his family and his duties in Ithaca . Tennyson projects this zeal into Ulysses ' unquenched desire for knowledge :
| [
{
"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 Tennyson 's character , however , is not the lover of public affairs seen in Homer 's poems . Rather , \" Ulisse \" from Dante 's Inferno is Tennyson 's main source for the character , which has an important effect on the poem 's interpretation . Ulisse recalls his voyage in the Inferno 's 26th canto , in which he is condemned to the Eighth Circle of false counsellors for misusing his gift of reason . Dante treats Ulisse , with his \" zeal … / T 'explore the world \" , as an evil counsellor who lusts for adventure at the expense of his family and his duties in Ithaca . Tennyson projects this zeal into Ulysses ' unquenched desire for knowledge : \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 poet 's intention to recall the Homeric character remains evident in certain passages . " I am become a name " ( 11 ) recalls an episode in the Odyssey in which Demodocus sings about Odysseus ' adventures in the king 's presence , acknowledging his fame . With phrases such as " There gloom the dark broad seas " ( 45 ) and " The deep / Moans round with many voices " ( 55 – 56 ) , Tennyson seems to be consciously invoking Homer .
| [
{
"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 poet 's intention to recall the Homeric character remains evident in certain passages . \" I am become a name \" ( 11 ) recalls an episode in the Odyssey in which Demodocus sings about Odysseus ' adventures in the king 's presence , acknowledging his fame . With phrases such as \" There gloom the dark broad seas \" ( 45 ) and \" The deep / Moans round with many voices \" ( 55 – 56 ) , Tennyson seems to be consciously invoking Homer . \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 poet's intention to recall the Homeric character remains evident in<|end-PI|> <|retrieval|> <|start-KFI|>certain passages<|end-KFI|>. <|start-PI|>\"I am become a name\" (11) recalls<|end-PI|> <|retrieval|> <|start-KFI|>an episode in the Odyssey in which Demodocus sings about Odysseus' adventures in the king's presence, acknowledging his fame<|end-KFI|>. <|start-PI|>With phrases such as \"There gloom the dark broad seas\" (45) and \"The deep / Moans round with many voices\" (55 – 56), Tennyson seems to be<|end-PI|> <|retrieval|> <|start-KFI|>consciously invoking Homer<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of the poet's intention, while the KFI indicates where this intention is evident. The <|retrieval|> token is inserted to indicate where additional information about these specific passages might be needed.",
"Ideal Knowledge": "A detailed analysis of the poem, highlighting specific passages that demonstrate the poet's intention to recall Homeric characters, with examples and explanations of the Homeric references.",
"Key Factual Information": "certain passages",
"Prefix reformulated": "How does the poet's intention to recall the Homeric character remain evident?",
"Prefix-intent": "The poet's intention to recall the Homeric character remains evident in",
"Verification": "The KFI directly answers how the poet's intention is evident, as prompted by the PI."
},
{
"Explanation": "The PI introduces a specific phrase from the poem, and the KFI provides the detailed Homeric reference it recalls. The <|retrieval|> token is inserted to indicate where additional information about this Homeric episode might be needed.",
"Ideal Knowledge": "A detailed explanation of the referenced episode in Homer's Odyssey, including the context of Demodocus singing about Odysseus, the significance of fame in Greek epic poetry, and how this relates to the phrase \"I am become a name.\"",
"Key Factual Information": "an episode in the Odyssey in which Demodocus sings about Odysseus' adventures in the king's presence, acknowledging his fame",
"Prefix reformulated": "What does the phrase \"I am become a name\" (11) recall in Homer's works?",
"Prefix-intent": "\"I am become a name\" (11) recalls",
"Verification": "The KFI directly provides the specific Homeric reference that the phrase recalls, as prompted by the PI."
},
{
"Explanation": "The PI provides specific phrases from Tennyson's work, setting up the question of their purpose, while the KFI explains that these are deliberate references to Homer. The <|retrieval|> token is inserted to indicate where additional information about Homer's style and its influence on Tennyson might be needed.",
"Ideal Knowledge": "An analysis of Homer's epic style, particularly his descriptions of the sea, and how Tennyson's phrases mirror or evoke this style. Examples from Homer's works that demonstrate similar imagery or language would be helpful.",
"Key Factual Information": "consciously invoking Homer",
"Prefix reformulated": "How does Tennyson use phrases like \"There gloom the dark broad seas\" (45) and \"The deep / Moans round with many voices\" (55 – 56)?",
"Prefix-intent": "With phrases such as \"There gloom the dark broad seas\" (45) and \"The deep / Moans round with many voices\" (55 – 56), Tennyson seems to be",
"Verification": "The KFI directly states the purpose of Tennyson's use of these phrases, as prompted by the PI."
}
]
} |
Critics have also noted the influence of Shakespeare in two passages . In the early movement , the savage race " That hoard , and sleep , and feed , and know not me " ( 5 ) echoes Hamlet 's soliloquy : " What is a man , / If his chief good and market of his time / Be but to sleep and feed ? A beast , no more . " Tennyson 's " How dull it is to pause , to make an end , / To rust unburnish ’ d , not to shine in use ! " ( 22 – 23 ) recalls Shakespeare 's Ulysses in Troilus and Cressida ( c . 1602 ) :
| [
{
"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 Critics have also noted the influence of Shakespeare in two passages . In the early movement , the savage race \" That hoard , and sleep , and feed , and know not me \" ( 5 ) echoes Hamlet 's soliloquy : \" What is a man , / If his chief good and market of his time / Be but to sleep and feed ? A beast , no more . \" Tennyson 's \" How dull it is to pause , to make an end , / To rust unburnish ’ d , not to shine in use ! \" ( 22 – 23 ) recalls Shakespeare 's Ulysses in Troilus and Cressida ( c . 1602 ) : \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|>Critics have also noted the influence of Shakespeare in two passages<|end-PI|> <|retrieval|> <|start-KFI|>In the early movement, the savage race 'That hoard, and sleep, and feed, and know not me' (5) echoes Hamlet's soliloquy: 'What is a man, / If his chief good and market of his time / Be but to sleep and feed? A beast, no more.'<|end-KFI|> <|start-PI|>Tennyson's 'How dull it is to pause, to make an end, / To rust unburnish'd, not to shine in use!' (22-23) recalls<|end-PI|> <|retrieval|> <|start-KFI|>Shakespeare's Ulysses in Troilus and Cressida (c. 1602)<|end-KFI|>:",
"none": null,
"results": [
{
"Explanation": "The PI introduces the topic of Shakespeare's influence, and the KFI provides the first specific example. The <|retrieval|> token is inserted to indicate where additional information about Shakespeare's influence might be needed.",
"Ideal Knowledge": "A detailed analysis of Tennyson's 'Ulysses' poem, highlighting specific lines and their connections to Shakespeare's works, particularly Hamlet. This should include a comparison of the themes and language used in both texts.",
"Key Factual Information": "In the early movement, the savage race 'That hoard, and sleep, and feed, and know not me' (5) echoes Hamlet's soliloquy: 'What is a man, / If his chief good and market of his time / Be but to sleep and feed? A beast, no more.'",
"Prefix reformulated": "What are the two passages where critics have noted Shakespeare's influence?",
"Prefix-intent": "Critics have also noted the influence of Shakespeare in two passages",
"Verification": "The KFI directly provides the first passage where Shakespeare's influence is noted, as prompted by the PI."
},
{
"Explanation": "The PI presents Tennyson's lines and sets up a query about their Shakespearean connection, while the KFI provides the specific character and play. The <|retrieval|> token is inserted to indicate where additional information about the connection between these works might be needed.",
"Ideal Knowledge": "A comparison between Tennyson's 'Ulysses' and Shakespeare's 'Troilus and Cressida', focusing on the character of Ulysses in both works. This should include an analysis of the similar themes and language used, particularly the idea of remaining active and useful in old age.",
"Key Factual Information": "Shakespeare's Ulysses in Troilus and Cressida (c. 1602)",
"Prefix reformulated": "Which Shakespeare character does Tennyson's line 'How dull it is to pause, to make an end, / To rust unburnish'd, not to shine in use!' recall?",
"Prefix-intent": "Tennyson's 'How dull it is to pause, to make an end, / To rust unburnish'd, not to shine in use!' (22-23) recalls",
"Verification": "The KFI directly identifies the Shakespeare character and play that Tennyson's lines recall, as prompted by the PI."
}
]
} |
The last movement of " Ulysses " , which is among the most familiar passages in nineteenth @-@ century English @-@ language poetry , presents decisive evidence of the influence of Dante . Ulysses turns his attention from himself and his kingdom and speaks of ports , seas , and his mariners . The strains of discontent and weakness in old age remain throughout the poem , but Tennyson finally leaves Ulysses " To strive , to seek , to find , and not to yield " ( 70 ) , recalling the Dantesque damnable desire for knowledge beyond all bounds . The words of Dante 's character as he exhorts his men to the journey find parallel in those of Tennyson 's Ulysses , who calls his men to join him on one last voyage . Quoting Dante 's Ulisse :
| [
{
"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 last movement of \" Ulysses \" , which is among the most familiar passages in nineteenth @-@ century English @-@ language poetry , presents decisive evidence of the influence of Dante . Ulysses turns his attention from himself and his kingdom and speaks of ports , seas , and his mariners . The strains of discontent and weakness in old age remain throughout the poem , but Tennyson finally leaves Ulysses \" To strive , to seek , to find , and not to yield \" ( 70 ) , recalling the Dantesque damnable desire for knowledge beyond all bounds . The words of Dante 's character as he exhorts his men to the journey find parallel in those of Tennyson 's Ulysses , who calls his men to join him on one last voyage . Quoting Dante 's Ulisse : \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
} |
However , critics note that in the Homeric narrative , Ulysses ' original mariners are dead . A significant irony therefore develops from Ulysses ' speech to his sailors — " Come , my friends , / ' Tis not too late to seek a newer world " ( 56 – 57 ) . Since Dante 's Ulisse has already undertaken this voyage and recounts it in the Inferno , Ulysses ' entire monologue can be envisioned as his recollection while situated in Hell .
| [
{
"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 However , critics note that in the Homeric narrative , Ulysses ' original mariners are dead . A significant irony therefore develops from Ulysses ' speech to his sailors — \" Come , my friends , / ' Tis not too late to seek a newer world \" ( 56 – 57 ) . Since Dante 's Ulisse has already undertaken this voyage and recounts it in the Inferno , Ulysses ' entire monologue can be envisioned as his recollection while situated in Hell . \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 degree to which Tennyson identifies with Ulysses has provided one of the great debates among scholars of the poem . Critics who find that Tennyson identifies with the speaker read Ulysses ' speech " affirmatively " , or without irony . Many other interpretations of the poem have developed from the argument that Tennyson does not identify with Ulysses , and further criticism has suggested that the purported inconsistencies in Ulysses ' character are the fault of the poet himself .
| [
{
"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 degree to which Tennyson identifies with Ulysses has provided one of the great debates among scholars of the poem . Critics who find that Tennyson identifies with the speaker read Ulysses ' speech \" affirmatively \" , or without irony . Many other interpretations of the poem have developed from the argument that Tennyson does not identify with Ulysses , and further criticism has suggested that the purported inconsistencies in Ulysses ' character are the fault of the poet himself . \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
} |
Key to the affirmative reading of " Ulysses " is the biographical context of the poem . Such a reading takes into account Tennyson 's statements about writing the poem — " the need of going forward " — and considers that he would not undermine Ulysses ' determination with irony when he needed a similar stalwartness to face life after Hallam 's death . Ulysses is thus seen as an heroic character whose determination to seek " some work of noble note " ( 52 ) is courageous in the face of a " still hearth " ( 2 ) and old age . The passion and conviction of Tennyson 's language — and even his own comments on the poem — signify that the poet , as was typical in the Victorian age , admired courage and persistence . Read straightforwardly , " Ulysses " promotes the questing spirit of youth , even in old age , and a refusal to resign and face life passively .
| [
{
"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 Key to the affirmative reading of \" Ulysses \" is the biographical context of the poem . Such a reading takes into account Tennyson 's statements about writing the poem — \" the need of going forward \" — and considers that he would not undermine Ulysses ' determination with irony when he needed a similar stalwartness to face life after Hallam 's death . Ulysses is thus seen as an heroic character whose determination to seek \" some work of noble note \" ( 52 ) is courageous in the face of a \" still hearth \" ( 2 ) and old age . The passion and conviction of Tennyson 's language — and even his own comments on the poem — signify that the poet , as was typical in the Victorian age , admired courage and persistence . Read straightforwardly , \" Ulysses \" promotes the questing spirit of youth , even in old age , and a refusal to resign and face life passively . \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
} |
Until the early twentieth century , readers reacted to " Ulysses " sympathetically . The meaning of the poem was increasingly debated as Tennyson 's stature rose . After Paull F. Baum criticized Ulysses ' inconsistencies and Tennyson 's conception of the poem in 1948 , the ironic interpretation became dominant . Baum finds in Ulysses echoes of Lord Byron 's flawed heroes , who similarly display conflicting emotions , self @-@ critical introspection , and a rejection of social responsibility . Even Ulysses ' resolute final utterance — " To strive , to seek , to find , and not to yield " — is undercut by irony , when Baum and later critics compare this line to Satan 's " courage never to submit or yield " in John Milton 's Paradise Lost ( 1667 ) .
| [
{
"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 Until the early twentieth century , readers reacted to \" Ulysses \" sympathetically . The meaning of the poem was increasingly debated as Tennyson 's stature rose . After Paull F. Baum criticized Ulysses ' inconsistencies and Tennyson 's conception of the poem in 1948 , the ironic interpretation became dominant . Baum finds in Ulysses echoes of Lord Byron 's flawed heroes , who similarly display conflicting emotions , self @-@ critical introspection , and a rejection of social responsibility . Even Ulysses ' resolute final utterance — \" To strive , to seek , to find , and not to yield \" — is undercut by irony , when Baum and later critics compare this line to Satan 's \" courage never to submit or yield \" in John Milton 's Paradise Lost ( 1667 ) . \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
} |
Ulysses ' apparent disdain for those around him is another facet of the ironic perspective . He declares that he is " matched with an aged wife " ( 3 ) , indicates his weariness in governing a " savage race " ( 4 ) , and suggests his philosophical distance from his son Telemachus . A skeptical reading of the second paragraph finds it a condescending tribute to Telemachus and a rejection of his " slow prudence " ( 36 ) . However , the adjectives used to describe Telemachus — " blameless " , " discerning " , and " decent " — are words with positive connotations in other of Tennyson 's poetry and within the classical tradition , where " blameless " is an attribute of gods and heroes .
| [
{
"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 Ulysses ' apparent disdain for those around him is another facet of the ironic perspective . He declares that he is \" matched with an aged wife \" ( 3 ) , indicates his weariness in governing a \" savage race \" ( 4 ) , and suggests his philosophical distance from his son Telemachus . A skeptical reading of the second paragraph finds it a condescending tribute to Telemachus and a rejection of his \" slow prudence \" ( 36 ) . However , the adjectives used to describe Telemachus — \" blameless \" , \" discerning \" , and \" decent \" — are words with positive connotations in other of Tennyson 's poetry and within the classical tradition , where \" blameless \" is an attribute of gods and heroes . \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
} |
Critic E. J. Chiasson argued in 1954 that Ulysses is without faith in an afterlife , and that Tennyson uses a " method of indirection " to affirm the need for religious faith by showing how Ulysses ' lack of faith leads to his neglect of kingdom and family . Chiasson regards the poem as " intractable " in Tennyson 's canon , but finds that the poem 's meaning resolves itself when this indirection is understood : it illustrates Tennyson 's conviction that " disregarding religious sanctions and ' submitting all things to desire ' leads to either a sybaritic or a brutal repudiation of responsibility and ' life ' . "
| [
{
"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 Critic E. J. Chiasson argued in 1954 that Ulysses is without faith in an afterlife , and that Tennyson uses a \" method of indirection \" to affirm the need for religious faith by showing how Ulysses ' lack of faith leads to his neglect of kingdom and family . Chiasson regards the poem as \" intractable \" in Tennyson 's canon , but finds that the poem 's meaning resolves itself when this indirection is understood : it illustrates Tennyson 's conviction that \" disregarding religious sanctions and ' submitting all things to desire ' leads to either a sybaritic or a brutal repudiation of responsibility and ' life ' . \" \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|>Critic E. J. Chiasson argued in 1954 that Ulysses is<|end-PI|> <|retrieval|> <|start-KFI|>without faith in an afterlife<|end-KFI|>, and that <|start-PI|>Tennyson uses a 'method of indirection' to affirm<|end-PI|> <|retrieval|> <|start-KFI|>the need for religious faith by showing how Ulysses' lack of faith leads to his neglect of kingdom and family<|end-KFI|>. <|start-PI|>Chiasson regards the poem as 'intractable' in Tennyson's canon, but finds that the poem's meaning resolves itself when<|end-PI|> <|retrieval|> <|start-KFI|>this indirection is understood<|end-KFI|>: <|start-PI|>it illustrates Tennyson's conviction that<|end-PI|> <|retrieval|> <|start-KFI|>'disregarding religious sanctions and 'submitting all things to desire' leads to either a sybaritic or a brutal repudiation of responsibility and 'life''<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces Chiasson's argument, and the KFI provides the specific claim about Ulysses' lack of faith. The <|retrieval|> token is inserted to indicate where additional information about Chiasson's critique might be needed.",
"Ideal Knowledge": "A summary of E. J. Chiasson's 1954 critique of Tennyson's 'Ulysses', focusing on the character's lack of faith in an afterlife and its implications for the poem's interpretation.",
"Key Factual Information": "without faith in an afterlife",
"Prefix reformulated": "What did critic E. J. Chiasson argue about Ulysses in 1954?",
"Prefix-intent": "Critic E. J. Chiasson argued in 1954 that Ulysses is",
"Verification": "The KFI directly provides Chiasson's argument about Ulysses as requested in the PI."
},
{
"Explanation": "The PI introduces Tennyson's technique, and the KFI provides the specific purpose and effect of this technique. The <|retrieval|> token is inserted to indicate where additional information about Tennyson's poetic methods might be needed.",
"Ideal Knowledge": "An analysis of Tennyson's poetic techniques, specifically the 'method of indirection' in 'Ulysses', and how it's used to convey themes of religious faith and responsibility.",
"Key Factual Information": "the need for religious faith by showing how Ulysses' lack of faith leads to his neglect of kingdom and family",
"Prefix reformulated": "What does Tennyson affirm using a 'method of indirection'?",
"Prefix-intent": "Tennyson uses a 'method of indirection' to affirm",
"Verification": "The KFI directly explains what Tennyson affirms through his method of indirection, as prompted by the PI."
},
{
"Explanation": "The PI sets up Chiasson's view of the poem and the question of its resolution, while the KFI provides the specific condition for understanding. The <|retrieval|> token is inserted to indicate where additional information about Chiasson's analysis might be needed.",
"Ideal Knowledge": "A detailed explanation of Chiasson's interpretation of 'Ulysses', focusing on how understanding Tennyson's indirect method clarifies the poem's meaning within his broader body of work.",
"Key Factual Information": "this indirection is understood",
"Prefix reformulated": "How does Chiasson find that the poem's meaning resolves itself?",
"Prefix-intent": "Chiasson regards the poem as 'intractable' in Tennyson's canon, but finds that the poem's meaning resolves itself when",
"Verification": "The KFI directly provides the condition for the poem's meaning to resolve itself, as prompted by the PI."
},
{
"Explanation": "The PI introduces Tennyson's conviction, and the KFI provides the specific belief illustrated by the poem. The <|retrieval|> token is inserted to indicate where additional information about Tennyson's philosophical views might be needed.",
"Ideal Knowledge": "An in-depth analysis of Tennyson's philosophical and moral views as expressed in 'Ulysses', particularly his ideas about the consequences of disregarding religious and moral principles.",
"Key Factual Information": "disregarding religious sanctions and 'submitting all things to desire' leads to either a sybaritic or a brutal repudiation of responsibility and 'life'",
"Prefix reformulated": "What conviction of Tennyson does the poem illustrate?",
"Prefix-intent": "it illustrates Tennyson's conviction that",
"Verification": "The KFI directly states Tennyson's conviction as illustrated by the poem, as prompted by the PI."
}
]
} |
Other ironic readings have found Ulysses longing for withdrawal , even death , in the form of his proposed quest . In noting the sense of passivity in the poem , critics highlight Tennyson 's tendency toward the melancholic . T. S. Eliot opines that " Tennyson could not tell a story at all " . He finds Dante 's treatment of Ulysses exciting , while Tennyson 's piece is " an elegiac mood " . " Ulysses " is found lacking in narrative action ; the hero 's goal is vague , and by the poem 's famous last line , it is not clear for what he is " striving " , or to what he refuses to yield . According to Victorian scholar Herbert Tucker , Tennyson 's characters " move " through time and space to be moved inwardly . To Ulysses , experience is " somewhere out there " ,
| [
{
"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 Other ironic readings have found Ulysses longing for withdrawal , even death , in the form of his proposed quest . In noting the sense of passivity in the poem , critics highlight Tennyson 's tendency toward the melancholic . T. S. Eliot opines that \" Tennyson could not tell a story at all \" . He finds Dante 's treatment of Ulysses exciting , while Tennyson 's piece is \" an elegiac mood \" . \" Ulysses \" is found lacking in narrative action ; the hero 's goal is vague , and by the poem 's famous last line , it is not clear for what he is \" striving \" , or to what he refuses to yield . According to Victorian scholar Herbert Tucker , Tennyson 's characters \" move \" through time and space to be moved inwardly . To Ulysses , experience is \" somewhere out there \" , \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 contemporary reviews of " Ulysses " were positive and found no irony in the poem . Author John Sterling — like Tennyson a member of the Cambridge Apostles — wrote in the Quarterly Review in 1842 , " How superior is ' Ulysses ' ! There is in this work a delightful epic tone , and a clear impassioned wisdom quietly carving its sage words and graceful figures on pale but lasting marble . " Tennyson 's 1842 volume of poetry impressed Scottish writer Thomas Carlyle . Quoting three lines of " Ulysses " in an 1842 letter to Tennyson —
| [
{
"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 contemporary reviews of \" Ulysses \" were positive and found no irony in the poem . Author John Sterling — like Tennyson a member of the Cambridge Apostles — wrote in the Quarterly Review in 1842 , \" How superior is ' Ulysses ' ! There is in this work a delightful epic tone , and a clear impassioned wisdom quietly carving its sage words and graceful figures on pale but lasting marble . \" Tennyson 's 1842 volume of poetry impressed Scottish writer Thomas Carlyle . Quoting three lines of \" Ulysses \" in an 1842 letter to Tennyson — \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
} |
— Carlyle remarked , " These lines do not make me weep , but there is in me what would fill whole Lachrymatories as I read . "
| [
{
"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 — Carlyle remarked , \" These lines do not make me weep , but there is in me what would fill whole Lachrymatories as I read . \" \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|>Carlyle remarked,<|end-PI|> <|retrieval|> <|start-KFI|>\"These lines do not make me weep, but there is in me what would fill whole Lachrymatories as I read.\"<|end-KFI|>",
"none": null,
"results": [
{
"Explanation": "The PI introduces Carlyle's remark, and the KFI provides his exact words. The <|retrieval|> token is inserted to indicate where additional context about Carlyle's literary opinions might be needed.",
"Ideal Knowledge": "Information about Carlyle's literary critiques and emotional responses to poetry, specifically his comments on the power of certain lines to evoke deep emotions without causing literal tears.",
"Key Factual Information": "\"These lines do not make me weep, but there is in me what would fill whole Lachrymatories as I read.\"",
"Prefix reformulated": "What did Carlyle remark about these lines?",
"Prefix-intent": "Carlyle remarked,",
"Verification": "The KFI directly provides Carlyle's remark, as prompted by the PI."
}
]
} |
English theologian Richard Holt Hutton summarized the poem as Tennyson 's " friendly picture of the insatiable craving for new experience , enterprise , and adventure , when under the control of a luminous reason and a self @-@ controlled will . " The contemporary poet Matthew Arnold was early in observing the narrative irony of the poem : he found Ulysses ' speech " the least plain , the most un @-@ Homeric , which can possibly be conceived . Homer presents his thought to you just as it wells from the source of his mind : Mr. Tennyson carefully distils his thought before he will part with it . Hence comes ... a heightened and elaborate air . "
| [
{
"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 English theologian Richard Holt Hutton summarized the poem as Tennyson 's \" friendly picture of the insatiable craving for new experience , enterprise , and adventure , when under the control of a luminous reason and a self @-@ controlled will . \" The contemporary poet Matthew Arnold was early in observing the narrative irony of the poem : he found Ulysses ' speech \" the least plain , the most un @-@ Homeric , which can possibly be conceived . Homer presents his thought to you just as it wells from the source of his mind : Mr. Tennyson carefully distils his thought before he will part with it . Hence comes ... a heightened and elaborate air . \" \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
} |
Despite the critical acclaim " Ulysses " received , its rise within the Tennyson canon took decades . Tennyson did not usually select it for publication in poetry anthologies ; in teaching anthologies , however , the poem was usually included — and it remains a popular teaching poem today . Its current prominence in Tennyson 's oeuvre is the result of two trends , according to Tennyson scholar Matthew Rowlinson : the rise of formal English poetry studies in the late nineteenth century , and the Victorian effort to articulate a British culture that could be exported . He argues that " Ulysses " forms part of the prehistory of imperialism — a term that only appeared in the language in 1851 . The protagonist sounds like a " colonial administrator " , and his reference to seeking a newer world ( 57 ) echoes the phrase " New World " , which became common during the Renaissance . While " Ulysses " cannot be read as overtly imperialistic , Tennyson 's later work as Poet Laureate sometimes argues for the value of Britain 's colonies , or was accused of jingoism . Rowlinson invokes the Marxist theorist Louis Althusser 's extension of the argument that ideology is ahistorical , finding that Tennyson 's poem " comes before an ideological construction for which it nonetheless makes people nostalgic " .
| [
{
"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 Despite the critical acclaim \" Ulysses \" received , its rise within the Tennyson canon took decades . Tennyson did not usually select it for publication in poetry anthologies ; in teaching anthologies , however , the poem was usually included — and it remains a popular teaching poem today . Its current prominence in Tennyson 's oeuvre is the result of two trends , according to Tennyson scholar Matthew Rowlinson : the rise of formal English poetry studies in the late nineteenth century , and the Victorian effort to articulate a British culture that could be exported . He argues that \" Ulysses \" forms part of the prehistory of imperialism — a term that only appeared in the language in 1851 . The protagonist sounds like a \" colonial administrator \" , and his reference to seeking a newer world ( 57 ) echoes the phrase \" New World \" , which became common during the Renaissance . While \" Ulysses \" cannot be read as overtly imperialistic , Tennyson 's later work as Poet Laureate sometimes argues for the value of Britain 's colonies , or was accused of jingoism . Rowlinson invokes the Marxist theorist Louis Althusser 's extension of the argument that ideology is ahistorical , finding that Tennyson 's poem \" comes before an ideological construction for which it nonetheless makes people nostalgic \" . \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": "Despite the critical acclaim \" Ulysses \" received , <|start-PI|> its rise within the Tennyson canon took <|end-PI|> <|retrieval|> <|start-KFI|> decades <|end-KFI|> . Tennyson did not usually select it for publication in poetry anthologies ; in teaching anthologies , however , the poem was usually included — and it remains a popular teaching poem today . <|start-PI|> Its current prominence in Tennyson 's oeuvre is the result of two trends , according to Tennyson scholar Matthew Rowlinson : <|end-PI|> <|retrieval|> <|start-KFI|> the rise of formal English poetry studies in the late nineteenth century , and the Victorian effort to articulate a British culture that could be exported <|end-KFI|> . <|start-PI|> He argues that \" Ulysses \" forms part of the prehistory of <|end-PI|> <|retrieval|> <|start-KFI|> imperialism — a term that only appeared in the language in 1851 <|end-KFI|> . The protagonist sounds like a \" colonial administrator \" , and his reference to seeking a newer world ( 57 ) echoes the phrase \" New World \" , which became common during the Renaissance . While \" Ulysses \" cannot be read as overtly imperialistic , Tennyson 's later work as Poet Laureate sometimes argues for the value of Britain 's colonies , or was accused of jingoism . <|start-PI|> Rowlinson invokes the Marxist theorist Louis Althusser 's extension of the argument that <|end-PI|> <|retrieval|> <|start-KFI|> ideology is ahistorical <|end-KFI|> , finding that Tennyson 's poem \" comes before an ideological construction for which it nonetheless makes people nostalgic \" .",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of \"Ulysses\" receiving critical acclaim, while the KFI provides the timeframe for its rise in importance. The <|retrieval|> token is inserted to indicate where additional information about the poem's reception history might be needed.",
"Ideal Knowledge": "A timeline of the reception and critical evaluation of Tennyson's works, focusing on the gradual rise in prominence of \"Ulysses\" over several decades.",
"Key Factual Information": "decades",
"Prefix reformulated": "How long did it take for \"Ulysses\" to rise within the Tennyson canon?",
"Prefix-intent": "Despite the critical acclaim \" Ulysses \" received , its rise within the Tennyson canon took",
"Verification": "The KFI directly answers the question posed by the reformulated PI about the time it took for \"Ulysses\" to gain prominence."
},
{
"Explanation": "The PI introduces Matthew Rowlinson's perspective on the reasons for the poem's prominence, while the KFI lists the two specific trends. The <|retrieval|> token is inserted to indicate where additional information about these trends might be needed.",
"Ideal Knowledge": "Detailed information on the development of English poetry studies in the late 19th century and the cultural context of Victorian Britain, particularly efforts to promote British culture abroad.",
"Key Factual Information": "the rise of formal English poetry studies in the late nineteenth century , and the Victorian effort to articulate a British culture that could be exported",
"Prefix reformulated": "What are the two trends that led to the current prominence of \"Ulysses\" in Tennyson's oeuvre, according to Matthew Rowlinson?",
"Prefix-intent": "Its current prominence in Tennyson 's oeuvre is the result of two trends , according to Tennyson scholar Matthew Rowlinson :",
"Verification": "The KFI directly provides the two trends mentioned in the PI, as stated by Matthew Rowlinson."
},
{
"Explanation": "The PI introduces Rowlinson's argument about \"Ulysses,\" while the KFI provides the specific concept (imperialism) and its historical context. The <|retrieval|> token is inserted to indicate where additional information about the relationship between literature and imperialism might be needed.",
"Ideal Knowledge": "Information on the historical context of imperialism, its emergence as a concept, and its relationship to literature in the mid-19th century, particularly in connection to Tennyson's works.",
"Key Factual Information": "imperialism — a term that only appeared in the language in 1851",
"Prefix reformulated": "What does Matthew Rowlinson argue that \"Ulysses\" forms part of the prehistory of?",
"Prefix-intent": "He argues that \" Ulysses \" forms part of the prehistory of",
"Verification": "The KFI directly provides Rowlinson's argument about \"Ulysses\" and its connection to imperialism, as prompted by the PI."
},
{
"Explanation": "The PI introduces Rowlinson's use of Althusser's theory, while the KFI provides the specific argument about ideology. The <|retrieval|> token is inserted to indicate where additional information about Althusser's theory might be needed.",
"Ideal Knowledge": "An explanation of Louis Althusser's Marxist theory, particularly his views on ideology and its relationship to history, and how this theory has been applied to literary analysis.",
"Key Factual Information": "ideology is ahistorical",
"Prefix reformulated": "What argument does Rowlinson invoke from Louis Althusser?",
"Prefix-intent": "Rowlinson invokes the Marxist theorist Louis Althusser 's extension of the argument that",
"Verification": "The KFI directly provides the argument from Althusser that Rowlinson invokes, as prompted by the PI."
}
]
} |
In a 1929 essay , T. S. Eliot called " Ulysses " a " perfect poem " . An analogue of Ulysses is found in Eliot 's " Gerontion " ( 1920 ) . Both poems are narrated by an aged man contemplating life 's end . An excerpt from " Gerontion " reads as an ironic comment on the introductory lines of " Ulysses " :
| [
{
"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 a 1929 essay , T. S. Eliot called \" Ulysses \" a \" perfect poem \" . An analogue of Ulysses is found in Eliot 's \" Gerontion \" ( 1920 ) . Both poems are narrated by an aged man contemplating life 's end . An excerpt from \" Gerontion \" reads as an ironic comment on the introductory lines of \" Ulysses \" : \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 Italian poet Giovanni Pascoli ( 1855 – 1912 ) stated that his long lyric poem L 'ultimo viaggio was an attempt to reconcile the portrayals of Ulysses in Dante and Tennyson with Tiresias 's prophecy that Ulysses would die " a mild death off the sea " . Pascoli 's Ulysses leaves Ithaca to retrace his epic voyage rather than begin another .
| [
{
"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 Italian poet Giovanni Pascoli ( 1855 – 1912 ) stated that his long lyric poem L 'ultimo viaggio was an attempt to reconcile the portrayals of Ulysses in Dante and Tennyson with Tiresias 's prophecy that Ulysses would die \" a mild death off the sea \" . Pascoli 's Ulysses leaves Ithaca to retrace his epic voyage rather than begin another . \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
} |
" Ulysses " remains much admired , even as the twentieth century brought new interpretations of the poem . Professor of literature Basil Willey commented in 1956 , " In ' Ulysses ' the sense that he must press on and not moulder in idleness is expressed objectively , through the classical story , and not subjectively as his own experience . [ Tennyson ] comes here as near perfection in the grand manner as he ever did ; the poem is flawless in tone from beginning to end ; spare , grave , free from excessive decoration , and full of firmly controlled feeling . " In the fifteenth edition of Bartlett 's Familiar Quotations ( 1980 ) , nine sections of " Ulysses " , comprising 36 of the poem 's 70 lines , are quoted , compared to only six in the ninth edition ( 1891 ) .
| [
{
"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 \" Ulysses \" remains much admired , even as the twentieth century brought new interpretations of the poem . Professor of literature Basil Willey commented in 1956 , \" In ' Ulysses ' the sense that he must press on and not moulder in idleness is expressed objectively , through the classical story , and not subjectively as his own experience . [ Tennyson ] comes here as near perfection in the grand manner as he ever did ; the poem is flawless in tone from beginning to end ; spare , grave , free from excessive decoration , and full of firmly controlled feeling . \" In the fifteenth edition of Bartlett 's Familiar Quotations ( 1980 ) , nine sections of \" Ulysses \" , comprising 36 of the poem 's 70 lines , are quoted , compared to only six in the ninth edition ( 1891 ) . \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
} |
Many readers have accepted the acclaimed last lines of " Ulysses " as inspirational . The poem 's ending line has also been used as a motto by schools and other organisations . U.S. Senator Robert Francis Kennedy quoted the three last lines at the end of his speech " On the Mindless Menace of Violence " in America a day after the assassination of Martin Luther King . The final line is inscribed on a cross at Observation Hill , Antarctica , to commemorate explorer Robert Falcon Scott and his party , who died on their return trek from the South Pole in 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 Many readers have accepted the acclaimed last lines of \" Ulysses \" as inspirational . The poem 's ending line has also been used as a motto by schools and other organisations . U.S. Senator Robert Francis Kennedy quoted the three last lines at the end of his speech \" On the Mindless Menace of Violence \" in America a day after the assassination of Martin Luther King . The final line is inscribed on a cross at Observation Hill , Antarctica , to commemorate explorer Robert Falcon Scott and his party , who died on their return trek from the South Pole in 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": null,
"none": "",
"results": null
} |
The Food Album is a compilation album by American singer @-@ songwriter " Weird Al " Yankovic , released on June 22 , 1993 by Scotti Brothers Records . The release features ten of Yankovic 's song parodies , all of which pertain to food . A similar album , The TV Album , which features songs entirely about television , would be released two years later .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n The Food Album is a compilation album by American singer @-@ songwriter \" Weird Al \" Yankovic , released on June 22 , 1993 by Scotti Brothers Records . The release features ten of Yankovic 's song parodies , all of which pertain to food . A similar album , The TV Album , which features songs entirely about television , would be released two years later . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
The album was begrudgingly released by Yankovic , who felt that the compilation was unnecessary and merely a way for his record label to make money . Several food @-@ related songs that Yankovic had recorded , such as " Girls Just Want to Have Lunch " and " Waffle King " were left off the record , although the former was due to personal preference , while the latter was due to scheduling 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 The album was begrudgingly released by Yankovic , who felt that the compilation was unnecessary and merely a way for his record label to make money . Several food @-@ related songs that Yankovic had recorded , such as \" Girls Just Want to Have Lunch \" and \" Waffle King \" were left off the record , although the former was due to personal preference , while the latter was due to scheduling 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
} |
The Food Album received mixed reviews from music critics , many of whom felt that the record was an enjoyable collection of songs , but that it was not an essential record to purchase . Despite the lukewarm reception , the record was certified Gold by the Recording Industry Association of America ( RIAA ) , making it Yankovic 's first and only compilation record to reach this certification .
| [
{
"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 Food Album received mixed reviews from music critics , many of whom felt that the record was an enjoyable collection of songs , but that it was not an essential record to purchase . Despite the lukewarm reception , the record was certified Gold by the Recording Industry Association of America ( RIAA ) , making it Yankovic 's first and only compilation record to reach this certification . \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 music featured on The Food Album spans a decade , with the earliest songs being recorded in 1982 , and the most recent song being recorded in 1992 . Yankovic 's first eponymous album has two songs featured : " I Love Rocky Road " and " My Bologna " . Both " Eat It " and " Theme from Rocky XIII ( The Rye or the Kaiser ) " were culled from the 1984 release , " Weird Al " Yankovic in 3 @-@ D. " Addicted to Spuds " originally appeared on the 1986 release Polka Party ! , and " Fat " and " Lasagna " were first featured on Yankovic 's 1988 release Even Worse . " Spam " first was released on the soundtrack to the 1989 film UHF . The final two songs — " The White Stuff " and " Taco Grande " — were taken from the 1992 album Off the Deep End .
| [
{
"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 music featured on The Food Album spans a decade , with the earliest songs being recorded in 1982 , and the most recent song being recorded in 1992 . Yankovic 's first eponymous album has two songs featured : \" I Love Rocky Road \" and \" My Bologna \" . Both \" Eat It \" and \" Theme from Rocky XIII ( The Rye or the Kaiser ) \" were culled from the 1984 release , \" Weird Al \" Yankovic in 3 @-@ D. \" Addicted to Spuds \" originally appeared on the 1986 release Polka Party ! , and \" Fat \" and \" Lasagna \" were first featured on Yankovic 's 1988 release Even Worse . \" Spam \" first was released on the soundtrack to the 1989 film UHF . The final two songs — \" The White Stuff \" and \" Taco Grande \" — were taken from the 1992 album Off the Deep End . \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
} |
Notable for its absence is " Girls Just Want to Have Lunch " , from Dare to Be Stupid ( 1985 ) , Yankovic 's only previously released food @-@ related song not to make the album . According to Yankovic , this is due to the fact there is a " royalty ceiling " on the albums and he needed to pick one song to cut from the list in order to turn a profit on the album . " Girls Just Want to Have Lunch " was chosen due to Yankovic 's personal dislike of the song , as his record label had forced him to record it in order to release Dare to Be Stupid back in 1985 . Also absent from the release is " Waffle King . " The song had originally been recorded for Off the Deep End . However , Yankovic decided to swap " Waffle King " with " I Was Only Kidding " — a song he had actually recorded for his next album — at the last minute ; this forced Yankovic to shelve " Waffle King " for the time being . The song was later released on " Smells Like Nirvana " single , as well as Yankovic 's eighth studio album , Alapalooza , which was released four months after The Food Album .
| [
{
"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 Notable for its absence is \" Girls Just Want to Have Lunch \" , from Dare to Be Stupid ( 1985 ) , Yankovic 's only previously released food @-@ related song not to make the album . According to Yankovic , this is due to the fact there is a \" royalty ceiling \" on the albums and he needed to pick one song to cut from the list in order to turn a profit on the album . \" Girls Just Want to Have Lunch \" was chosen due to Yankovic 's personal dislike of the song , as his record label had forced him to record it in order to release Dare to Be Stupid back in 1985 . Also absent from the release is \" Waffle King . \" The song had originally been recorded for Off the Deep End . However , Yankovic decided to swap \" Waffle King \" with \" I Was Only Kidding \" — a song he had actually recorded for his next album — at the last minute ; this forced Yankovic to shelve \" Waffle King \" for the time being . The song was later released on \" Smells Like Nirvana \" single , as well as Yankovic 's eighth studio album , Alapalooza , which was released four months after The Food Album . \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 album was released by Scotti Brothers Records and was only begrudgingly approved by Yankovic . At the time , Scotti Brothers had insisted on putting out a new album by Yankovic in order to meet monetary projections for the fiscal quarter , despite the fact that no new album was ready ; Alapalooza would not be released until later in the year . The original concept was to release a record entitled Al Unplugged , which would have featured a cover depicting Yankovic holding the cords for unplugged kitchen appliances , but instead of being a live album featuring live performances , it would have featured studio remixes of previously released material , with the electronic instruments missing . Yankovic convinced them to release The Food Album instead — " a concept [ he ] hated only slightly less " — but would later describe it as a " cheesy compilation " put out " against [ his ] better wishes and judgement . "
| [
{
"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 album was released by Scotti Brothers Records and was only begrudgingly approved by Yankovic . At the time , Scotti Brothers had insisted on putting out a new album by Yankovic in order to meet monetary projections for the fiscal quarter , despite the fact that no new album was ready ; Alapalooza would not be released until later in the year . The original concept was to release a record entitled Al Unplugged , which would have featured a cover depicting Yankovic holding the cords for unplugged kitchen appliances , but instead of being a live album featuring live performances , it would have featured studio remixes of previously released material , with the electronic instruments missing . Yankovic convinced them to release The Food Album instead — \" a concept [ he ] hated only slightly less \" — but would later describe it as a \" cheesy compilation \" put out \" against [ his ] better wishes and judgement . \" \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 TV Album was released under similar circumstances in 1995 ; however , when it came time to release the latter album , Yankovic reported that " the record company was a whole lot nicer when they asked the second time " , and that there was " more groveling [ and ] less demanding " . Following the release of The Food Album and The TV Album — in addition to the various greatest hits records that had been released — Scotti Brothers used @-@ up all of their compilation options in Yankovic 's contract , which prevented the release of further compilations when Volcano Records acquired his contract in the late 1990s .
| [
{
"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 TV Album was released under similar circumstances in 1995 ; however , when it came time to release the latter album , Yankovic reported that \" the record company was a whole lot nicer when they asked the second time \" , and that there was \" more groveling [ and ] less demanding \" . Following the release of The Food Album and The TV Album — in addition to the various greatest hits records that had been released — Scotti Brothers used @-@ up all of their compilation options in Yankovic 's contract , which prevented the release of further compilations when Volcano Records acquired his contract in the late 1990s . \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 TV Album was released under similar circumstances in<|end-PI|> <|retrieval|> <|start-KFI|>1995<|end-KFI|> ; however , <|start-PI|>when it came time to release the latter album , Yankovic reported that<|end-PI|> <|retrieval|> <|start-KFI|>\"the record company was a whole lot nicer when they asked the second time\" , and that there was \"more groveling [ and ] less demanding\"<|end-KFI|> . <|start-PI|>Following the release of The Food Album and The TV Album — in addition to the various greatest hits records that had been released — Scotti Brothers<|end-PI|> <|retrieval|> <|start-KFI|>used @-@ up all of their compilation options in Yankovic 's contract , which prevented the release of further compilations when Volcano Records acquired his contract in the late 1990s<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces The TV Album and sets up a query about its release date, while the KFI provides the specific year. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "A chronological list of Weird Al Yankovic's album releases, including compilation albums, with specific focus on The TV Album and its release date in 1995.",
"Key Factual Information": "1995",
"Prefix reformulated": "When was The TV Album released?",
"Prefix-intent": "The TV Album was released under similar circumstances in",
"Verification": "The KFI directly provides the year of release as requested in the PI."
},
{
"Explanation": "The PI sets up Yankovic's report about the album release, and the KFI provides his direct quotes. The <|retrieval|> token is inserted to indicate where additional context about Yankovic's relationship with record companies might be needed.",
"Ideal Knowledge": "Detailed accounts of Weird Al Yankovic's interactions with record companies, specifically regarding the release of compilation albums like The TV Album, including direct quotes from Yankovic about the process.",
"Key Factual Information": "\"the record company was a whole lot nicer when they asked the second time\" , and that there was \"more groveling [ and ] less demanding\"",
"Prefix reformulated": "What did Weird Al Yankovic report about the release of the latter album?",
"Prefix-intent": "when it came time to release the latter album , Yankovic reported that",
"Verification": "The KFI directly quotes Yankovic's report about the record company's behavior, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of multiple album releases, and the KFI explains the contractual consequences. The <|retrieval|> token is inserted to indicate where additional information about record label contracts and compilation rights might be needed.",
"Ideal Knowledge": "Detailed information about Weird Al Yankovic's contract with Scotti Brothers, including specifics about compilation album clauses, and how these affected his later contract with Volcano Records.",
"Key Factual Information": "used @-@ up all of their compilation options in Yankovic 's contract , which prevented the release of further compilations when Volcano Records acquired his contract in the late 1990s",
"Prefix reformulated": "What did Scotti Brothers do after releasing The Food Album, The TV Album, and various greatest hits records?",
"Prefix-intent": "Following the release of The Food Album and The TV Album — in addition to the various greatest hits records that had been released — Scotti Brothers",
"Verification": "The KFI directly explains the consequences of Scotti Brothers' actions regarding Yankovic's contract, as prompted by the PI."
}
]
} |
The album artwork — which features a cartoon alien after it has eaten Yankovic — was created by Doug Lawrence , who is better known as " Mr. Lawrence " , an American voice actor , comedian , writer , storyboard artist , animator and director . The " grotesque " cover was Yankovic 's " passive @-@ aggressive protest " against his label for forcing out the album ; Yankovic intended the alien having " picked the desiccated corpse of Weird Al clean " to be a reference to his record label " bleed [ ing ] his catalogue dry " by releasing the album . The Japanese release of the album , however , featured much different artwork , as well as a name change ; because there is no " F " in the Japanese language , the album was retitled The Hood .
| [
{
"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 album artwork — which features a cartoon alien after it has eaten Yankovic — was created by Doug Lawrence , who is better known as \" Mr. Lawrence \" , an American voice actor , comedian , writer , storyboard artist , animator and director . The \" grotesque \" cover was Yankovic 's \" passive @-@ aggressive protest \" against his label for forcing out the album ; Yankovic intended the alien having \" picked the desiccated corpse of Weird Al clean \" to be a reference to his record label \" bleed [ ing ] his catalogue dry \" by releasing the album . The Japanese release of the album , however , featured much different artwork , as well as a name change ; because there is no \" F \" in the Japanese language , the album was retitled The Hood . \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 album artwork — which features a cartoon alien after it has eaten Yankovic — was created by<|end-PI|> <|retrieval|> <|start-KFI|>Doug Lawrence , who is better known as \" Mr. Lawrence \" , an American voice actor , comedian , writer , storyboard artist , animator and director<|end-KFI|> . <|start-PI|>The \" grotesque \" cover was Yankovic 's<|end-PI|> <|retrieval|> <|start-KFI|>\" passive @-@ aggressive protest \" against his label for forcing out the album<|end-KFI|> ; Yankovic intended the alien having \" picked the desiccated corpse of Weird Al clean \" to be a reference to his record label \" bleed [ ing ] his catalogue dry \" by releasing the album . <|start-PI|>The Japanese release of the album , however , featured<|end-PI|> <|retrieval|> <|start-KFI|>much different artwork , as well as a name change ; because there is no \" F \" in the Japanese language , the album was retitled The Hood<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of the album artwork and asks who created it, while the KFI provides the artist's name and background. The <|retrieval|> token is inserted to indicate where additional information about the artist might be needed.",
"Ideal Knowledge": "Biographical information about Doug Lawrence (Mr. Lawrence), including his career as an artist and his work on album covers, particularly for Weird Al Yankovic.",
"Key Factual Information": "Doug Lawrence , who is better known as \" Mr. Lawrence \" , an American voice actor , comedian , writer , storyboard artist , animator and director",
"Prefix reformulated": "Who created the album artwork featuring a cartoon alien eating Yankovic?",
"Prefix-intent": "The album artwork — which features a cartoon alien after it has eaten Yankovic — was created by",
"Verification": "The KFI directly provides the creator's name and details, answering the question posed in the PI."
},
{
"Explanation": "The PI describes the cover as 'grotesque' and sets up a query about its purpose, while the KFI provides the specific reason behind the design choice. The <|retrieval|> token is inserted to indicate where additional context about Yankovic's relationship with his label might be needed.",
"Ideal Knowledge": "Information about Weird Al Yankovic's relationship with his record label, including any conflicts or disagreements about album releases and artistic decisions.",
"Key Factual Information": "\" passive @-@ aggressive protest \" against his label for forcing out the album",
"Prefix reformulated": "What was the purpose of Yankovic's 'grotesque' album cover?",
"Prefix-intent": "The \" grotesque \" cover was Yankovic 's",
"Verification": "The KFI directly explains the purpose of the album cover design, as prompted by the PI."
},
{
"Explanation": "The PI introduces the Japanese release and sets up a query about its differences, while the KFI provides specific details about the changes made. The <|retrieval|> token is inserted to indicate where additional information about international album releases might be needed.",
"Ideal Knowledge": "Information about the localization of album releases in different countries, particularly focusing on changes made to Weird Al Yankovic's albums for the Japanese market, including artwork modifications and title adaptations.",
"Key Factual Information": "much different artwork , as well as a name change ; because there is no \" F \" in the Japanese language , the album was retitled The Hood",
"Prefix reformulated": "What was different about the Japanese release of Yankovic's album?",
"Prefix-intent": "The Japanese release of the album , however , featured",
"Verification": "The KFI directly describes the differences in the Japanese release, answering the question posed in the PI."
}
]
} |
The Food Album has received mixed reviews from most critics ; many felt that while the album was amusing it was not an essential release . Allmusic reviewer Johnny Loftus awarded the album three out of five stars and wrote that , " The Food Album is an enjoyable bag of treats . Just don 't eat too much , or you 'll probably get sick . " Likewise , The Rolling Stone Album Guide awarded the album three out of five stars . Anthony Violanti of The Buffalo News gave the album a moderately positive review and wrote that , " [ t ] here are two kinds of people in the world : those who love Weird Al Yankovic and those who can 't stand him . Count me among the Weird One 's biggest fans , and that 's why I flipped out when listening to The Food Album . " He concluded that the album was " like reading Mad magazine " ; he gave the record three stars out of five . Tim Grobaty of the Press @-@ Telegram , on the other hand , wrote negatively of the album , stating " Yankovic 's songs are the kinds of things that are sort of funny in concept , less funny when you actually hear them once , and increasingly irritating with each subsequent listen [ and ] his food songs are among his worst . "
| [
{
"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 Food Album has received mixed reviews from most critics ; many felt that while the album was amusing it was not an essential release . Allmusic reviewer Johnny Loftus awarded the album three out of five stars and wrote that , \" The Food Album is an enjoyable bag of treats . Just don 't eat too much , or you 'll probably get sick . \" Likewise , The Rolling Stone Album Guide awarded the album three out of five stars . Anthony Violanti of The Buffalo News gave the album a moderately positive review and wrote that , \" [ t ] here are two kinds of people in the world : those who love Weird Al Yankovic and those who can 't stand him . Count me among the Weird One 's biggest fans , and that 's why I flipped out when listening to The Food Album . \" He concluded that the album was \" like reading Mad magazine \" ; he gave the record three stars out of five . Tim Grobaty of the Press @-@ Telegram , on the other hand , wrote negatively of the album , stating \" Yankovic 's songs are the kinds of things that are sort of funny in concept , less funny when you actually hear them once , and increasingly irritating with each subsequent listen [ and ] his food songs are among his worst . \" \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
} |
Upon release , The Food Album failed to chart ; however , it sold steadily . On January 25 , 2006 — more than ten years after its release — the album was certified Gold by the Recording Industry Association of America ( RIAA ) . This makes it Yankovic 's first and only compilation album to sell over 500 @,@ 000 copies and be certified Gold .
| [
{
"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 Upon release , The Food Album failed to chart ; however , it sold steadily . On January 25 , 2006 — more than ten years after its release — the album was certified Gold by the Recording Industry Association of America ( RIAA ) . This makes it Yankovic 's first and only compilation album to sell over 500 @,@ 000 copies and be certified Gold . \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 Patriarchal Cathedral of the Holy Ascension of God ( Bulgarian : Патриаршеска катедрала „ Свето Възнесение Господне “ , Patriarsheska katedrala „ Sveto Vaznesenie Gospodne “ ) is a former Eastern Orthodox cathedral in the city of Veliko Tarnovo , in north central Bulgaria . Located on top of the fortified Tsarevets hill in the former capital of the Second Bulgarian Empire , the cathedral was the seat of the Bulgarian patriarch from its construction in the 11th – 12th century to its destruction in 1393 .
| [
{
"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 Patriarchal Cathedral of the Holy Ascension of God ( Bulgarian : Патриаршеска катедрала „ Свето Възнесение Господне “ , Patriarsheska katedrala „ Sveto Vaznesenie Gospodne “ ) is a former Eastern Orthodox cathedral in the city of Veliko Tarnovo , in north central Bulgaria . Located on top of the fortified Tsarevets hill in the former capital of the Second Bulgarian Empire , the cathedral was the seat of the Bulgarian patriarch from its construction in the 11th – 12th century to its destruction in 1393 . \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
} |
Standing on top of a late Roman church , the cathedral , reconstructed in the 1970s and 1980s , follows a cross @-@ domed plan with a bell tower and a triple apse . Richly decorated on both the exterior and interior , its internal walls now feature modern frescoes , the presence of which has meant that it has not been reconsecrated . Though not active as a Christian place of worship , it has been open for visitors since 1985 .
| [
{
"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 Standing on top of a late Roman church , the cathedral , reconstructed in the 1970s and 1980s , follows a cross @-@ domed plan with a bell tower and a triple apse . Richly decorated on both the exterior and interior , its internal walls now feature modern frescoes , the presence of which has meant that it has not been reconsecrated . Though not active as a Christian place of worship , it has been open for visitors since 1985 . \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 Patriarchal Cathedral of the Holy Ascension of God is not the first church building to occupy the position on top of the Tsarevets hill . It was constructed directly on top of a late Roman ( early Byzantine ) basilica which dates to the 5th – 6th century AD . The Roman basilica may have remained in use by the local congregation during the First Bulgarian Empire , though it was no longer active by the time the construction of the current church began .
| [
{
"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 Patriarchal Cathedral of the Holy Ascension of God is not the first church building to occupy the position on top of the Tsarevets hill . It was constructed directly on top of a late Roman ( early Byzantine ) basilica which dates to the 5th – 6th century AD . The Roman basilica may have remained in use by the local congregation during the First Bulgarian Empire , though it was no longer active by the time the construction of the current church began . \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 Patriarchal Cathedral of the Holy Ascension of God is not the first church building to occupy the position on top of the Tsarevets hill<|end-PI|> <|retrieval|> <|start-KFI|>It was constructed directly on top of a late Roman ( early Byzantine ) basilica which dates to the 5th – 6th century AD<|end-KFI|>. <|start-PI|>The Roman basilica may have remained in use by the local congregation during<|end-PI|> <|retrieval|> <|start-KFI|>the First Bulgarian Empire<|end-KFI|>, though it was no longer active by the time the construction of the current church began.",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context that the current cathedral is not the first structure on the site, and the KFI provides specific information about the previous building. The <|retrieval|> token is inserted to indicate where additional historical information might be needed.",
"Ideal Knowledge": "Detailed historical information about the structures that existed on Tsarevets hill before the Patriarchal Cathedral of the Holy Ascension of God, including dates, architectural styles, and historical context of the Roman/Byzantine basilica.",
"Key Factual Information": "It was constructed directly on top of a late Roman ( early Byzantine ) basilica which dates to the 5th – 6th century AD",
"Prefix reformulated": "What was the previous structure on the site of the Patriarchal Cathedral of the Holy Ascension of God on Tsarevets hill?",
"Prefix-intent": "The Patriarchal Cathedral of the Holy Ascension of God is not the first church building to occupy the position on top of the Tsarevets hill",
"Verification": "The KFI directly provides information about the previous structure on the site, as prompted by the PI."
},
{
"Explanation": "The PI introduces the possibility of continued use of the Roman basilica, and the KFI specifies the historical period. The <|retrieval|> token is inserted to indicate where additional information about the First Bulgarian Empire might be needed.",
"Ideal Knowledge": "Information about the First Bulgarian Empire, including its time period, religious practices, and the use of existing Roman structures during this era.",
"Key Factual Information": "the First Bulgarian Empire",
"Prefix reformulated": "During which period may the Roman basilica have remained in use by the local congregation?",
"Prefix-intent": "The Roman basilica may have remained in use by the local congregation during",
"Verification": "The KFI directly provides the time period during which the basilica may have been in use, as prompted by the PI."
}
]
} |
The current building of the Patriarchal Cathedral is considered by scholars to have been built in two stages . The first stage of construction was carried out in the late 11th century or the 12th century . The cathedral was initially built as a monastery church in the middle of a monastery compound , though in the early 12th century it was already the seat of the Bulgarian patriarch . The compound suffered large @-@ scale damage caused by a fire , which necessitated the church 's reconstruction in middle of the 14th century , perhaps during the rule of Tsar Ivan Alexander of Bulgaria ( r . 1331 – 71 ) . Besides repair and reinforcement efforts , work on the church in the 14th century also included the construction of the exonarthex and the bell tower .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n The current building of the Patriarchal Cathedral is considered by scholars to have been built in two stages . The first stage of construction was carried out in the late 11th century or the 12th century . The cathedral was initially built as a monastery church in the middle of a monastery compound , though in the early 12th century it was already the seat of the Bulgarian patriarch . The compound suffered large @-@ scale damage caused by a fire , which necessitated the church 's reconstruction in middle of the 14th century , perhaps during the rule of Tsar Ivan Alexander of Bulgaria ( r . 1331 – 71 ) . Besides repair and reinforcement efforts , work on the church in the 14th century also included the construction of the exonarthex and the bell tower . \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 current building of the Patriarchal Cathedral is considered by scholars to have been built in<|end-PI|> <|retrieval|> <|start-KFI|>two stages<|end-KFI|>. <|start-PI|>The first stage of construction was carried out in<|end-PI|> <|retrieval|> <|start-KFI|>the late 11th century or the 12th century<|end-KFI|>. <|start-PI|>The cathedral was initially built as a monastery church in the middle of a monastery compound , though in the early 12th century it was already<|end-PI|> <|retrieval|> <|start-KFI|>the seat of the Bulgarian patriarch<|end-KFI|>. <|start-PI|>The compound suffered large @-@ scale damage caused by a fire , which necessitated the church 's reconstruction in<|end-PI|> <|retrieval|> <|start-KFI|>middle of the 14th century , perhaps during the rule of Tsar Ivan Alexander of Bulgaria ( r . 1331 – 71 )<|end-KFI|>. Besides repair and reinforcement efforts , work on the church in the 14th century also included the construction of the exonarthex and the bell tower.",
"none": null,
"results": [
{
"Explanation": "The PI introduces the topic of the cathedral's construction, and the KFI provides the specific number of stages. The <|retrieval|> token is inserted to indicate where additional information about the construction stages might be needed.",
"Ideal Knowledge": "A detailed historical account of the construction phases of the Patriarchal Cathedral, including scholarly consensus on the number of major building stages and their approximate time periods.",
"Key Factual Information": "two stages",
"Prefix reformulated": "In how many stages was the current building of the Patriarchal Cathedral built according to scholars?",
"Prefix-intent": "The current building of the Patriarchal Cathedral is considered by scholars to have been built in",
"Verification": "The KFI directly answers the question posed in the reformulated PI about the number of stages in which the cathedral was built."
},
{
"Explanation": "The PI sets up a query about the timing of the first construction stage, and the KFI provides the specific time period. The <|retrieval|> token is inserted to indicate where additional information about the construction timeline might be needed.",
"Ideal Knowledge": "A timeline of the Patriarchal Cathedral's construction, focusing on the initial building phase with specific dates or century ranges, supported by archaeological and historical evidence.",
"Key Factual Information": "the late 11th century or the 12th century",
"Prefix reformulated": "When was the first stage of construction of the Patriarchal Cathedral carried out?",
"Prefix-intent": "The first stage of construction was carried out in",
"Verification": "The KFI directly provides the time period for the first stage of construction, as prompted by the PI."
},
{
"Explanation": "The PI provides context about the cathedral's initial purpose and sets up a query about its function in the early 12th century, while the KFI provides the specific role it held. The <|retrieval|> token is inserted to indicate where additional information about the cathedral's changing functions might be needed.",
"Ideal Knowledge": "A historical account of the Patriarchal Cathedral's changing roles and functions, specifically detailing its transition from a monastery church to the seat of the Bulgarian patriarch, with dates and contextual information about the Bulgarian Orthodox Church's hierarchy.",
"Key Factual Information": "the seat of the Bulgarian patriarch",
"Prefix reformulated": "What was the function of the Patriarchal Cathedral in the early 12th century?",
"Prefix-intent": "The cathedral was initially built as a monastery church in the middle of a monastery compound , though in the early 12th century it was already",
"Verification": "The KFI directly states the function of the cathedral in the early 12th century, as prompted by the PI."
},
{
"Explanation": "The PI introduces the fire damage and sets up a query about the timing of reconstruction, while the KFI provides the specific time period and historical context. The <|retrieval|> token is inserted to indicate where additional information about the reconstruction and the reign of Tsar Ivan Alexander might be needed.",
"Ideal Knowledge": "A detailed account of the fire that damaged the Patriarchal Cathedral, including its date and extent of damage, followed by information about the reconstruction efforts in the 14th century, with specific focus on the reign of Tsar Ivan Alexander and his role in the cathedral's restoration.",
"Key Factual Information": "middle of the 14th century , perhaps during the rule of Tsar Ivan Alexander of Bulgaria ( r . 1331 – 71 )",
"Prefix reformulated": "When was the Patriarchal Cathedral reconstructed due to fire damage?",
"Prefix-intent": "The compound suffered large @-@ scale damage caused by a fire , which necessitated the church 's reconstruction in",
"Verification": "The KFI directly provides the time period of the reconstruction and additional context about the ruler, as prompted by the PI."
}
]
} |
There are several references to the cathedral in medieval sources . The earliest reference to the church tells of the transfer of Saint Michael the Warrior 's relics from the Potuka fortress to the Patriarchal Cathedral on the order of Tsar Kaloyan ( r . 1197 – 1207 ) . The housing of a warrior saint 's relics in the Patriarchal Cathedral signifies the incessant warfare against Byzantines and Latins that dominated Kaloyan 's reign . In the late 14th century , the last Patriarch of Tarnovo , Saint Evtimiy , described the church as the " great patriarch 's Cathedral of the Holy Ascension " in his writings .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n There are several references to the cathedral in medieval sources . The earliest reference to the church tells of the transfer of Saint Michael the Warrior 's relics from the Potuka fortress to the Patriarchal Cathedral on the order of Tsar Kaloyan ( r . 1197 – 1207 ) . The housing of a warrior saint 's relics in the Patriarchal Cathedral signifies the incessant warfare against Byzantines and Latins that dominated Kaloyan 's reign . In the late 14th century , the last Patriarch of Tarnovo , Saint Evtimiy , described the church as the \" great patriarch 's Cathedral of the Holy Ascension \" in his writings . \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
} |
Another possible reference to the church may be in a marginal note from 1358 to a copy of the Acts of the Apostles . In the note , the copyist , one Laloe , thanks God and the " Holy and Most Glorious Ascension " for having finished his work on the book . Scholar Bistra Nikolova believes this to be an allusion to the Patriarchal Cathedral , which may have patronised the project . Alternatively , the copy could have been made at the cathedral 's scriptorium , where Laloe may have worked .
| [
{
"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 Another possible reference to the church may be in a marginal note from 1358 to a copy of the Acts of the Apostles . In the note , the copyist , one Laloe , thanks God and the \" Holy and Most Glorious Ascension \" for having finished his work on the book . Scholar Bistra Nikolova believes this to be an allusion to the Patriarchal Cathedral , which may have patronised the project . Alternatively , the copy could have been made at the cathedral 's scriptorium , where Laloe may have worked . \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": "Another possible reference to the church may be in a marginal note from 1358 to a copy of the Acts of the Apostles . In the note , <|start-PI|> the copyist , one Laloe , thanks God and <|end-PI|> <|retrieval|> <|start-KFI|> the \" Holy and Most Glorious Ascension \" for having finished his work on the book <|end-KFI|> . <|start-PI|> Scholar Bistra Nikolova believes this to be <|end-PI|> <|retrieval|> <|start-KFI|> an allusion to the Patriarchal Cathedral , which may have patronised the project <|end-KFI|> . <|start-PI|> Alternatively , the copy could have been made <|end-PI|> <|retrieval|> <|start-KFI|> at the cathedral 's scriptorium , where Laloe may have worked <|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of a marginal note and introduces the copyist Laloe, while the KFI provides the specific entity that Laloe thanks. The <|retrieval|> token is inserted to indicate where additional information about medieval manuscript annotations might be needed.",
"Ideal Knowledge": "Information about medieval manuscript annotations, specifically those referencing religious entities or events, and their significance in identifying the provenance or context of the manuscript.",
"Key Factual Information": "the \" Holy and Most Glorious Ascension \" for having finished his work on the book",
"Prefix reformulated": "What does the copyist Laloe thank in the marginal note from 1358?",
"Prefix-intent": "Another possible reference to the church may be in a marginal note from 1358 to a copy of the Acts of the Apostles . In the note , the copyist , one Laloe , thanks God and",
"Verification": "The KFI directly provides the information about what Laloe thanks, as prompted by the PI."
},
{
"Explanation": "The PI introduces Scholar Bistra Nikolova's belief, while the KFI provides her specific interpretation. The <|retrieval|> token is inserted to indicate where additional information about scholarly interpretations of medieval manuscripts might be needed.",
"Ideal Knowledge": "Scholarly interpretations of medieval manuscript annotations, particularly those that link references to specific churches or cathedrals, and information about the Patriarchal Cathedral's role in manuscript production.",
"Key Factual Information": "an allusion to the Patriarchal Cathedral , which may have patronised the project",
"Prefix reformulated": "What does Scholar Bistra Nikolova believe the reference to 'Holy and Most Glorious Ascension' to be?",
"Prefix-intent": "Scholar Bistra Nikolova believes this to be",
"Verification": "The KFI directly provides Bistra Nikolova's interpretation of the reference, as prompted by the PI."
},
{
"Explanation": "The PI introduces an alternative possibility for the manuscript's origin, while the KFI provides the specific location and context. The <|retrieval|> token is inserted to indicate where additional information about medieval scriptoria might be needed.",
"Ideal Knowledge": "Information about medieval scriptoria, particularly those associated with cathedrals, and their role in manuscript production. Details about the organization and functioning of scriptoria would be helpful.",
"Key Factual Information": "at the cathedral 's scriptorium , where Laloe may have worked",
"Prefix reformulated": "Where else could the copy of the Acts of the Apostles have been made?",
"Prefix-intent": "Alternatively , the copy could have been made",
"Verification": "The KFI directly provides an alternative location for the manuscript's creation, as prompted by the PI."
}
]
} |
The church is also depicted in the medieval sketch of Tarnovo in the Braşov Menaion , a menaion service book written in the mid @-@ 14th century and then carried to Kronstadt ( now Braşov , Romania ) after the fall of Bulgaria under Ottoman rule .
| [
{
"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 church is also depicted in the medieval sketch of Tarnovo in the Braşov Menaion , a menaion service book written in the mid @-@ 14th century and then carried to Kronstadt ( now Braşov , Romania ) after the fall of Bulgaria under Ottoman rule . \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 Patriarchal Cathedral was destroyed after the Ottomans captured the Bulgarian capital after their Siege of Tarnovo on 17 July 1393 . The church was fully reconstructed in the 20th century ; reconstruction works were carried out by a team under architect Boyan Kuzupov . These commenced in 1978 and were finished in 1981 , to mark Bulgaria 's 1300th anniversary . However , it was not until November 1985 , when the contemporary murals were finished , that the church was opened once again for visitors . The church 's ruins have been protected as a national antiquity since 1927 ; in 1967 , they were proclaimed an architectural monument of culture of national importance . As part of the Tsarevets architectural reserve , it is also listed among the 100 Tourist Sites of Bulgaria .
| [
{
"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 Patriarchal Cathedral was destroyed after the Ottomans captured the Bulgarian capital after their Siege of Tarnovo on 17 July 1393 . The church was fully reconstructed in the 20th century ; reconstruction works were carried out by a team under architect Boyan Kuzupov . These commenced in 1978 and were finished in 1981 , to mark Bulgaria 's 1300th anniversary . However , it was not until November 1985 , when the contemporary murals were finished , that the church was opened once again for visitors . The church 's ruins have been protected as a national antiquity since 1927 ; in 1967 , they were proclaimed an architectural monument of culture of national importance . As part of the Tsarevets architectural reserve , it is also listed among the 100 Tourist Sites of Bulgaria . \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 Patriarchal Cathedral of the Holy Ascension of God is located on top of the Tsarevets hill , overlooking the modern city of Veliko Tarnovo . The church was part of a group of buildings which constituted the seat of the Bulgarian Patriarchate and acted as the city and the country 's main cathedral . The patriarchate on Tsarevets was a fortress of its own , with two defensive towers and an entrance on its west wall . The Patriarchal Cathedral stood in the middle of its courtyard .
| [
{
"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 Patriarchal Cathedral of the Holy Ascension of God is located on top of the Tsarevets hill , overlooking the modern city of Veliko Tarnovo . The church was part of a group of buildings which constituted the seat of the Bulgarian Patriarchate and acted as the city and the country 's main cathedral . The patriarchate on Tsarevets was a fortress of its own , with two defensive towers and an entrance on its west wall . The Patriarchal Cathedral stood in the middle of its courtyard . \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 Patriarchal Cathedral features a triple apse , the central part of which matches the apse of the original basilica on the site . The three @-@ naved church follows the traditional Byzantine cross @-@ in @-@ square design . Built out of crushed stones and mortar with limited brickwork , it measures 26 by 12 metres ( 85 ft × 39 ft ) . The cathedral includes two narthices , a bell tower and two other premises attached to the south church wall . The presence of a bell tower is considered to be a rarity in Balkan church architecture . Six columns support the interior and distinguish the altar from the cella ( naos ) . It is unclear whether the church housed a synthronon ( stone benches for the clergy ) in the apse , as there are doubts that its remains may actually be part of the older basilica .
| [
{
"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 Patriarchal Cathedral features a triple apse , the central part of which matches the apse of the original basilica on the site . The three @-@ naved church follows the traditional Byzantine cross @-@ in @-@ square design . Built out of crushed stones and mortar with limited brickwork , it measures 26 by 12 metres ( 85 ft × 39 ft ) . The cathedral includes two narthices , a bell tower and two other premises attached to the south church wall . The presence of a bell tower is considered to be a rarity in Balkan church architecture . Six columns support the interior and distinguish the altar from the cella ( naos ) . It is unclear whether the church housed a synthronon ( stone benches for the clergy ) in the apse , as there are doubts that its remains may actually be part of the older basilica . \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 church featured ample exterior and interior decoration . While the facades were decorated with arches and ceramic tiles , the interior floor mosaics were made of white , yellow and pink marble as well as semi @-@ precious gemstones like sapphire and porphyry . The interior walls were covered with frescoes and mosaics . However , none of the interior decoration has survived . During the church 's 20th @-@ century reconstruction , its interior was repainted by artist Teofan Sokerov , who depicted important moments of medieval Bulgarian history in a modernist style . Due to these murals , the church has never been reconsecrated and remains inactive . The facade of the cathedral also includes a stone with a donor 's inscription of a Bulgarian ruler , which ended up as part of the building material .
| [
{
"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 church featured ample exterior and interior decoration . While the facades were decorated with arches and ceramic tiles , the interior floor mosaics were made of white , yellow and pink marble as well as semi @-@ precious gemstones like sapphire and porphyry . The interior walls were covered with frescoes and mosaics . However , none of the interior decoration has survived . During the church 's 20th @-@ century reconstruction , its interior was repainted by artist Teofan Sokerov , who depicted important moments of medieval Bulgarian history in a modernist style . Due to these murals , the church has never been reconsecrated and remains inactive . The facade of the cathedral also includes a stone with a donor 's inscription of a Bulgarian ruler , which ended up as part of the building material . \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 church featured ample exterior and interior decoration . <|start-PI|> While the facades were decorated with <|end-PI|> <|retrieval|> <|start-KFI|> arches and ceramic tiles <|end-KFI|> , <|start-PI|> the interior floor mosaics were made of <|end-PI|> <|retrieval|> <|start-KFI|> white , yellow and pink marble as well as semi @-@ precious gemstones like sapphire and porphyry <|end-KFI|> . The interior walls were covered with frescoes and mosaics . However , none of the interior decoration has survived . <|start-PI|> During the church 's 20th @-@ century reconstruction , its interior was repainted by <|end-PI|> <|retrieval|> <|start-KFI|> artist Teofan Sokerov , who depicted important moments of medieval Bulgarian history in a modernist style <|end-KFI|> . <|start-PI|> Due to these murals , the church has <|end-PI|> <|retrieval|> <|start-KFI|> never been reconsecrated and remains inactive <|end-KFI|> . The facade of the cathedral also includes a stone with a donor 's inscription of a Bulgarian ruler , which ended up as part of the building material .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the church's decoration and sets up a query about the facades, while the KFI provides specific details. The <|retrieval|> token is inserted to indicate where additional information about church facade decorations might be needed.",
"Ideal Knowledge": "Detailed information about common architectural decorations used in medieval Bulgarian churches, specifically focusing on facade decorations including arches and ceramic tiles.",
"Key Factual Information": "arches and ceramic tiles",
"Prefix reformulated": "How were the facades of the church decorated?",
"Prefix-intent": "The church featured ample exterior and interior decoration . While the facades were decorated with",
"Verification": "The KFI directly provides the decoration details of the facades as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the materials used in the floor mosaics, and the KFI provides a detailed list. The <|retrieval|> token is inserted to indicate where additional information about mosaic materials might be needed.",
"Ideal Knowledge": "A comprehensive list of materials commonly used in medieval church floor mosaics, including various types of marble and semi-precious stones, with specific emphasis on white, yellow, and pink marble, sapphire, and porphyry.",
"Key Factual Information": "white , yellow and pink marble as well as semi @-@ precious gemstones like sapphire and porphyry",
"Prefix reformulated": "What materials were used for the church's interior floor mosaics?",
"Prefix-intent": "the interior floor mosaics were made of",
"Verification": "The KFI directly lists the materials used in the floor mosaics, as prompted by the PI."
},
{
"Explanation": "The PI introduces the church's reconstruction and sets up a query about who repainted it, while the KFI provides the artist's name and details about his work. The <|retrieval|> token is inserted to indicate where additional information about the artist might be needed.",
"Ideal Knowledge": "Biographical information about Teofan Sokerov, including his artistic style and notable works, particularly his contributions to the reconstruction of Bulgarian churches in the 20th century.",
"Key Factual Information": "artist Teofan Sokerov , who depicted important moments of medieval Bulgarian history in a modernist style",
"Prefix reformulated": "Who repainted the church's interior during its 20th-century reconstruction?",
"Prefix-intent": "During the church 's 20th @-@ century reconstruction , its interior was repainted by",
"Verification": "The KFI directly provides the name of the artist and describes his work, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the consequences of the murals, and the KFI provides the specific outcome. The <|retrieval|> token is inserted to indicate where additional information about church consecration rules might be needed.",
"Ideal Knowledge": "Information about the consecration requirements for Orthodox churches, particularly in relation to interior decorations and modernist art, and how these requirements affect the active status of a church.",
"Key Factual Information": "never been reconsecrated and remains inactive",
"Prefix reformulated": "What is the current status of the church due to Teofan Sokerov's murals?",
"Prefix-intent": "Due to these murals , the church has",
"Verification": "The KFI directly states the current status of the church as a result of the murals, as prompted by the PI."
}
]
} |
There are a total of four burial grounds in and around the church , two of which are burials for priests . One of the burial grounds is inside the exonarthex , where overarched tombs were built in the 14th century . Besides Michael the Warrior 's relics , the cathedral also housed the remains of Bulgarian patriarchs Joachim I , Macarius and Joachim III .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n There are a total of four burial grounds in and around the church , two of which are burials for priests . One of the burial grounds is inside the exonarthex , where overarched tombs were built in the 14th century . Besides Michael the Warrior 's relics , the cathedral also housed the remains of Bulgarian patriarchs Joachim I , Macarius and Joachim III . \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
} |
Daydream is the fifth studio album by American singer and songwriter Mariah Carey , released on October 3 , 1995 , by Columbia Records . The follow @-@ up to her internationally successful album Music Box ( 1993 ) and holiday album Merry Christmas ( 1994 ) , Daydream differed from the two by leaning increasingly towards R & B and hip hop . Throughout the project , Carey collaborated with Walter Afanasieff , with whom she wrote and produced most of her two previous albums . With Daydream , Carey took more control over the musical direction as well as the album 's composition . Carey said she considered Daydream the beginning of her musical and vocal transformation , a change that became more apparent in her sixth album Butterfly ( 1997 ) . During the album 's production , Carey endured many creative differences with her label and husband Tommy Mottola .
| [
{
"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 Daydream is the fifth studio album by American singer and songwriter Mariah Carey , released on October 3 , 1995 , by Columbia Records . The follow @-@ up to her internationally successful album Music Box ( 1993 ) and holiday album Merry Christmas ( 1994 ) , Daydream differed from the two by leaning increasingly towards R & B and hip hop . Throughout the project , Carey collaborated with Walter Afanasieff , with whom she wrote and produced most of her two previous albums . With Daydream , Carey took more control over the musical direction as well as the album 's composition . Carey said she considered Daydream the beginning of her musical and vocal transformation , a change that became more apparent in her sixth album Butterfly ( 1997 ) . During the album 's production , Carey endured many creative differences with her label and husband Tommy Mottola . \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 Daydream , Carey collaborated with Jermaine Dupri for the first time , and co @-@ wrote and produced a song with Kenneth " Babyface " Edmonds , with whom she had collaborated on Music Box . It was also the first time she had worked with Boyz II Men , an R & B group consisting of four male vocalists . Together , they wrote the concept and lyrics for " One Sweet Day , " a song that Carey co @-@ produced with Afanasieff . With his assistance and the addition of a few contemporary producers , she was able to make a subtle transition into the R & B market . Daydream was nominated for six Grammy Awards at the 38th annual ceremony , during which Carey performed live . Due to the album 's critical and commercial success , critics believed Carey would be one of the night 's big winners . However , to her dismay , she was completely shut out , causing the subject to become very public and controversial . She left the annual ceremony empty @-@ handed .
| [
{
"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 Daydream , Carey collaborated with Jermaine Dupri for the first time , and co @-@ wrote and produced a song with Kenneth \" Babyface \" Edmonds , with whom she had collaborated on Music Box . It was also the first time she had worked with Boyz II Men , an R & B group consisting of four male vocalists . Together , they wrote the concept and lyrics for \" One Sweet Day , \" a song that Carey co @-@ produced with Afanasieff . With his assistance and the addition of a few contemporary producers , she was able to make a subtle transition into the R & B market . Daydream was nominated for six Grammy Awards at the 38th annual ceremony , during which Carey performed live . Due to the album 's critical and commercial success , critics believed Carey would be one of the night 's big winners . However , to her dismay , she was completely shut out , causing the subject to become very public and controversial . She left the annual ceremony empty @-@ handed . \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
} |
Six singles were released from the album . Its lead single " Fantasy " became the first single by a female artist to debut at number one on the US Billboard Hot 100 and went on to top the chart for eight weeks and became the second best @-@ selling single of 1995 in the country . The song topped the charts in Australia , Canada , and New Zealand and became a top @-@ five hit in Finland , France , and the United Kingdom . The second single " One Sweet Day " topped the Billboard Hot 100 for sixteen weeks and became the longest @-@ running number one single in American history , a record it still holds . It also topped the charts in Canada and New Zealand and peaked within the top five in Australia , France , Ireland and the Netherlands . Jointly , the singles from Daydream spent a combined six months at the top of the Hot 100 . To promote Daydream , Carey embarked on the short but successful , Daydream World Tour , visiting Japan and Europe .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Six singles were released from the album . Its lead single \" Fantasy \" became the first single by a female artist to debut at number one on the US Billboard Hot 100 and went on to top the chart for eight weeks and became the second best @-@ selling single of 1995 in the country . The song topped the charts in Australia , Canada , and New Zealand and became a top @-@ five hit in Finland , France , and the United Kingdom . The second single \" One Sweet Day \" topped the Billboard Hot 100 for sixteen weeks and became the longest @-@ running number one single in American history , a record it still holds . It also topped the charts in Canada and New Zealand and peaked within the top five in Australia , France , Ireland and the Netherlands . Jointly , the singles from Daydream spent a combined six months at the top of the Hot 100 . To promote Daydream , Carey embarked on the short but successful , Daydream World Tour , visiting Japan and Europe . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
At the time of its release , Daydream became Carey 's best @-@ reviewed album . Critics universally praised her matured lyrics and songwriting , as well as her musical direction . The album became an international success , debuting at number one in nine different countries , and in the top five in almost every major music market . Daydream became Carey 's second album to be certified diamond by the Recording Industry Association of America ( RIAA ) , denoting shipments of ten million copies in the United States . Aside from its success in the United States , the album made the top five of the best @-@ selling albums in Japan by a non @-@ Asian artist , with 2 @.@ 1 million copies sold . Daydream remains one of the best @-@ selling albums of all time , with 25 million copies sold 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 At the time of its release , Daydream became Carey 's best @-@ reviewed album . Critics universally praised her matured lyrics and songwriting , as well as her musical direction . The album became an international success , debuting at number one in nine different countries , and in the top five in almost every major music market . Daydream became Carey 's second album to be certified diamond by the Recording Industry Association of America ( RIAA ) , denoting shipments of ten million copies in the United States . Aside from its success in the United States , the album made the top five of the best @-@ selling albums in Japan by a non @-@ Asian artist , with 2 @.@ 1 million copies sold . Daydream remains one of the best @-@ selling albums of all time , with 25 million copies sold 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
} |
Aside from being Carey 's second highest worldwide seller , Daydream served as her most personal and directly influenced album at the time . During the album 's recording , Carey grew as an artist , as well as a writer . For the first time in her career , Carey was able to make music that she truly related to , R & B and hip hop . While Columbia allowed Carey more leniency with the music she recorded , they became hesitant when she featured Ol ' Dirty # # # # # # # in the remix for " Fantasy . " They feared the sudden change was completely left field for her music , and worried it would jeopardize the album 's success . In an interview with Entertainment Weekly , Carey openly spoke of her issues with Columbia : " Everybody was like ' What , are you crazy ? ' They 're nervous about breaking the formula . It works to have me sing a ballad on stage in a long dress with my hair 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 Aside from being Carey 's second highest worldwide seller , Daydream served as her most personal and directly influenced album at the time . During the album 's recording , Carey grew as an artist , as well as a writer . For the first time in her career , Carey was able to make music that she truly related to , R & B and hip hop . While Columbia allowed Carey more leniency with the music she recorded , they became hesitant when she featured Ol ' Dirty # # # # # # # in the remix for \" Fantasy . \" They feared the sudden change was completely left field for her music , and worried it would jeopardize the album 's success . In an interview with Entertainment Weekly , Carey openly spoke of her issues with Columbia : \" Everybody was like ' What , are you crazy ? ' They 're nervous about breaking the formula . It works to have me sing a ballad on stage in a long dress with my hair 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
} |
While Carey 's new musical direction caused tension between her and Columbia , it began to severely strain her relationship with her husband at the time , Tommy Mottola . Mottola had always been involved in Carey 's career , because he was the head of Sony Music , the parent company of her label . Since the time of Carey 's debut , Mottola had controlled nearly every aspect of her career , keeping her sound carefully regulated and insisting that she continue recording middle @-@ of @-@ the @-@ road pop music , despite her interest in hip hop .
| [
{
"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 Carey 's new musical direction caused tension between her and Columbia , it began to severely strain her relationship with her husband at the time , Tommy Mottola . Mottola had always been involved in Carey 's career , because he was the head of Sony Music , the parent company of her label . Since the time of Carey 's debut , Mottola had controlled nearly every aspect of her career , keeping her sound carefully regulated and insisting that she continue recording middle @-@ of @-@ the @-@ road pop music , despite her interest in hip hop . \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|>While Carey's new musical direction caused tension between her and Columbia, it began to severely strain her relationship with<|end-PI|> <|retrieval|> <|start-KFI|>her husband at the time, Tommy Mottola<|end-KFI|>. <|start-PI|>Mottola had always been involved in Carey's career, because he was<|end-PI|> <|retrieval|> <|start-KFI|>the head of Sony Music, the parent company of her label<|end-KFI|>. <|start-PI|>Since the time of Carey's debut, Mottola had controlled nearly every aspect of her career, keeping her sound carefully regulated and insisting that she continue recording<|end-PI|> <|retrieval|> <|start-KFI|>middle-of-the-road pop music, despite her interest in hip hop<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of Carey's changing musical direction and its impact on her relationships, while the KFI identifies the specific personal relationship affected. The <|retrieval|> token is inserted to indicate where additional information about Tommy Mottola might be needed.",
"Ideal Knowledge": "Information about Mariah Carey's marriage to Tommy Mottola during the time of her musical direction change, including details about their personal and professional relationship.",
"Key Factual Information": "her husband at the time, Tommy Mottola",
"Prefix reformulated": "How did Mariah Carey's new musical direction affect her relationship with her husband?",
"Prefix-intent": "While Carey's new musical direction caused tension between her and Columbia, it began to severely strain her relationship with",
"Verification": "The KFI directly identifies the person whose relationship with Carey was strained, as prompted by the PI."
},
{
"Explanation": "The PI introduces Mottola's involvement in Carey's career, and the KFI provides the specific reason for this involvement. The <|retrieval|> token is inserted to indicate where additional information about Sony Music and its relationship to Carey's label might be needed.",
"Ideal Knowledge": "Detailed information about Tommy Mottola's role at Sony Music, including his position as head of the company and how it related to Mariah Carey's record label.",
"Key Factual Information": "the head of Sony Music, the parent company of her label",
"Prefix reformulated": "What was Tommy Mottola's position that allowed him to be involved in Mariah Carey's career?",
"Prefix-intent": "Mottola had always been involved in Carey's career, because he was",
"Verification": "The KFI directly provides Mottola's position and its relevance to Carey's career, as prompted by the PI."
},
{
"Explanation": "The PI describes Mottola's control over Carey's career and sets up a query about the type of music he insisted on, while the KFI provides the specific genre and contrasts it with Carey's interests. The <|retrieval|> token is inserted to indicate where additional information about musical genres and industry trends might be needed.",
"Ideal Knowledge": "Information about the musical styles of Mariah Carey's early career, including the contrast between middle-of-the-road pop and hip hop, and how these styles were perceived in the music industry at the time.",
"Key Factual Information": "middle-of-the-road pop music, despite her interest in hip hop",
"Prefix reformulated": "What type of music did Tommy Mottola insist Mariah Carey continue recording?",
"Prefix-intent": "Since the time of Carey's debut, Mottola had controlled nearly every aspect of her career, keeping her sound carefully regulated and insisting that she continue recording",
"Verification": "The KFI directly specifies the type of music Mottola insisted on and contrasts it with Carey's interests, as prompted by the PI."
}
]
} |
Carey confessed that she never tried to change the situation because " [ she ] used to be insecure and cautious , and so [ she ] would listen to what the people said . " However , the control Mottola exerted over her career soon " spilled into her personal life " once they were married , increasing the amount of conflict between the two . Soon , it was obvious that their marriage was in shambles ; as stated in a Vanity Fair article , " the couple began to argue at the drop of a hat . " Carey was very involved in the project , more so than she had ever been on an album . " I went into this phase of recording , recording , recording and doing it really fast , " she told Time . " This time , I had more time , and I focused more on what I wanted to do . " As Carey 's career and work continued to reflect her views on how it should sound , her marriage to Mottola continued to " deteriorate . "
| [
{
"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 Carey confessed that she never tried to change the situation because \" [ she ] used to be insecure and cautious , and so [ she ] would listen to what the people said . \" However , the control Mottola exerted over her career soon \" spilled into her personal life \" once they were married , increasing the amount of conflict between the two . Soon , it was obvious that their marriage was in shambles ; as stated in a Vanity Fair article , \" the couple began to argue at the drop of a hat . \" Carey was very involved in the project , more so than she had ever been on an album . \" I went into this phase of recording , recording , recording and doing it really fast , \" she told Time . \" This time , I had more time , and I focused more on what I wanted to do . \" As Carey 's career and work continued to reflect her views on how it should sound , her marriage to Mottola continued to \" deteriorate . \" \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 first songs that was recorded for the album was " Fantasy . " While Carey began developing new ideas for Daydream , she thought of the song " Genius of Love " by Tom Tom Club . She had always been a fan of the song , and presented Dave Hall with the idea of sampling the song 's hook . Hall incorporated a groove that he felt complimented Carey 's voice , while she composed some of the other beats and wrote the lyrics . Carey recorded a remix to the song as well , featuring hip @-@ hop verses from O.D.B of the Wu @-@ Tang Clan , as well as production from Puffy . She spoke highly of the remix , complimenting Puffy and O.D.B , " He 's so known in the street , and he 's one of the best people out there ... we kind of did what we both do and having O.D.B took it to another level . He was my ultimate choice , so I was really happy with the way it turned out . " " One Sweet Day " was a song that Carey wrote with the R & B group Boyz II Men . After Carey 's friend and past collaborator David Cole died , she began writing and developing a song that would pay homage to him and all the friends and family her fans had lost along life 's journey . Carey had the chorus and concept composed , and after meeting with Boyz II Men , they realized they too had a similar idea in development . Together , using Carey 's chorus and idea , as well as the melody they had produced , they wrote and composed the song . It was produced by Afanasieff , who built on the song 's melody and added various grooves and beats . Carey expressed how the song was " meant to be " and how all the pieces fit into place :
| [
{
"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 first songs that was recorded for the album was \" Fantasy . \" While Carey began developing new ideas for Daydream , she thought of the song \" Genius of Love \" by Tom Tom Club . She had always been a fan of the song , and presented Dave Hall with the idea of sampling the song 's hook . Hall incorporated a groove that he felt complimented Carey 's voice , while she composed some of the other beats and wrote the lyrics . Carey recorded a remix to the song as well , featuring hip @-@ hop verses from O.D.B of the Wu @-@ Tang Clan , as well as production from Puffy . She spoke highly of the remix , complimenting Puffy and O.D.B , \" He 's so known in the street , and he 's one of the best people out there ... we kind of did what we both do and having O.D.B took it to another level . He was my ultimate choice , so I was really happy with the way it turned out . \" \" One Sweet Day \" was a song that Carey wrote with the R & B group Boyz II Men . After Carey 's friend and past collaborator David Cole died , she began writing and developing a song that would pay homage to him and all the friends and family her fans had lost along life 's journey . Carey had the chorus and concept composed , and after meeting with Boyz II Men , they realized they too had a similar idea in development . Together , using Carey 's chorus and idea , as well as the melody they had produced , they wrote and composed the song . It was produced by Afanasieff , who built on the song 's melody and added various grooves and beats . Carey expressed how the song was \" meant to be \" and how all the pieces fit into place : \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
} |
I wrote the initial idea for ' One Sweet Day ' with Walter , and I had the chorus ... and I stopped and said , ' I really wanna do this with Boyz II Men , ' because ... obviously I 'm a big fan of theirs and I just thought that the work was crying out for them , the vocals that they do , so I put it away and said , ' Who knows if this could ever happen , but I just don 't wanna finish this song because I want it to be our song if we ever do it together . [ The ] whole idea of when you lose people that are close to you , it changes your life and changes your perspective . When they came into the studio , I played them the idea for the song and when [ it ] finished , they looked at each other , a bit stunned , and told me that Nat " Nathan Morris " had written a song for his road manager who had passed away . It had basically the same lyrics and fit over the same chord changes . It was really , really weird , we finished the song right then and there . We were all kinda flipped about it ourselves . Fate had a lot to do with that . I know some people won 't believe it , but we wouldn 't make up such a crazy story .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n I wrote the initial idea for ' One Sweet Day ' with Walter , and I had the chorus ... and I stopped and said , ' I really wanna do this with Boyz II Men , ' because ... obviously I 'm a big fan of theirs and I just thought that the work was crying out for them , the vocals that they do , so I put it away and said , ' Who knows if this could ever happen , but I just don 't wanna finish this song because I want it to be our song if we ever do it together . [ The ] whole idea of when you lose people that are close to you , it changes your life and changes your perspective . When they came into the studio , I played them the idea for the song and when [ it ] finished , they looked at each other , a bit stunned , and told me that Nat \" Nathan Morris \" had written a song for his road manager who had passed away . It had basically the same lyrics and fit over the same chord changes . It was really , really weird , we finished the song right then and there . We were all kinda flipped about it ourselves . Fate had a lot to do with that . I know some people won 't believe it , but we wouldn 't make up such a crazy story . \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
} |
While the album 's development was underway , Carey expressed interest in working with Jermaine Dupri , whom she had been a fan of since his 1992 song , " Jump . " Soon after , Carey , Dupri , and Manuel Seal began composing a song for the album . As Seal played the piano , Carey began humming and playing with certain notes in the B @-@ section , until she came up with the chorus for " Always Be My Baby . " After the rest of the song was written and composed , Carey recorded the song alongside longtime background singers Kelly Price , Shanrae Price , and Melonie Daniels . Together , they built " a wall of background voices " in which she would cover with her final belting notes . The song featured a downbeat rhythm , while its composition was described as " sassy and soft R & B " which displayed a " sexy and slow jam . " " Underneath the Stars " was the first song recorded for Daydream . The song featured a " ' 70s soul vibe " as well as synthetic record scratches , in order to the give the song an authentic ' 70s sound . Carey felt the additions were simple steps taken to further display a contemporary R & B groove . Additionally , she felt the song paid homage to the style of Minnie Riperton , who was one of Carey 's biggest vocal influences growing up . The song had a soft sound , and had " a lot [ sic ] of texture " and bass , showing a more creative side to Carey .
| [
{
"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 the album 's development was underway , Carey expressed interest in working with Jermaine Dupri , whom she had been a fan of since his 1992 song , \" Jump . \" Soon after , Carey , Dupri , and Manuel Seal began composing a song for the album . As Seal played the piano , Carey began humming and playing with certain notes in the B @-@ section , until she came up with the chorus for \" Always Be My Baby . \" After the rest of the song was written and composed , Carey recorded the song alongside longtime background singers Kelly Price , Shanrae Price , and Melonie Daniels . Together , they built \" a wall of background voices \" in which she would cover with her final belting notes . The song featured a downbeat rhythm , while its composition was described as \" sassy and soft R & B \" which displayed a \" sexy and slow jam . \" \" Underneath the Stars \" was the first song recorded for Daydream . The song featured a \" ' 70s soul vibe \" as well as synthetic record scratches , in order to the give the song an authentic ' 70s sound . Carey felt the additions were simple steps taken to further display a contemporary R & B groove . Additionally , she felt the song paid homage to the style of Minnie Riperton , who was one of Carey 's biggest vocal influences growing up . The song had a soft sound , and had \" a lot [ sic ] of texture \" and bass , showing a more creative side to Carey . \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
} |
For the album , Carey covered the 1982 Journey song " Open Arms . " The song was of Carey 's personal choice , as well as her own idea . Together with Afanasieff , they toned down the song 's arrangement , making it a bit glossy , especially in comparison to the " raw and powerful ' One Sweet Day . ' " Additionally , with the help of her background singers , Carey added a touch of gospel to the song . One of the more gospel @-@ influenced songs on the album was " I Am Free . " The song was created by Carey , Afanasieff and Loris Holland , with whom she had worked previously on Merry Christmas . Carey began humming the melody with the lyrics she had already written , while Holland played the organ and Afanasieff worked on the song 's programming . , giving the song a genuine and unforced gospel feel . The chorus was sophisticated and natural , with each following line " cascading onto one another , " something that would have proved difficult for a " less skilled vocalist . " Carey started leaning away from the " standard Celine Dion ballad " and more towards R & B jams . However , she was not going to completely abandon the type of songs that made her famous . For this reason , Carey wrote " When I Saw You " with Afanasieff , a song that would truly embody some of her earlier work , as well as show off her powerful vocals . Returning to her R & B territory , Carey recorded " Long Ago " , the second song she wrote alongside Dupri and Seal which contains a strong hip hop background . Her vocals in the song were described as " sliding over the insistent bassline like silk . "
| [
{
"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 For the album , Carey covered the 1982 Journey song \" Open Arms . \" The song was of Carey 's personal choice , as well as her own idea . Together with Afanasieff , they toned down the song 's arrangement , making it a bit glossy , especially in comparison to the \" raw and powerful ' One Sweet Day . ' \" Additionally , with the help of her background singers , Carey added a touch of gospel to the song . One of the more gospel @-@ influenced songs on the album was \" I Am Free . \" The song was created by Carey , Afanasieff and Loris Holland , with whom she had worked previously on Merry Christmas . Carey began humming the melody with the lyrics she had already written , while Holland played the organ and Afanasieff worked on the song 's programming . , giving the song a genuine and unforced gospel feel . The chorus was sophisticated and natural , with each following line \" cascading onto one another , \" something that would have proved difficult for a \" less skilled vocalist . \" Carey started leaning away from the \" standard Celine Dion ballad \" and more towards R & B jams . However , she was not going to completely abandon the type of songs that made her famous . For this reason , Carey wrote \" When I Saw You \" with Afanasieff , a song that would truly embody some of her earlier work , as well as show off her powerful vocals . Returning to her R & B territory , Carey recorded \" Long Ago \" , the second song she wrote alongside Dupri and Seal which contains a strong hip hop background . Her vocals in the song were described as \" sliding over the insistent bassline like silk . \" \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
} |
" Melt Away " was a song Carey produced on her own , and co @-@ wrote with Babyface . The song 's writing and production were " superb . " with each verse gliding into its chorus . According to Chris Nickson , " Underneath the Stars " was as " strong as any slow jam released in the nineties , and one that would find a lot of flavor late at night with dancers . " Another song that brought back reminders of older decades was " Forever . " The throwback was featured through the chord changes and in the way the guitar arpeggios " stayed at the forefront of the music . " The song displayed subtle vocals from Carey , as well as an undeniable richness . " Daydream Interlude ( Sweet Fantasy Dub Mix ) " was one of the liveliest tracks on the album . The song was a club remix of " Fantasy " , which was tuned and remixed by famed house music producer David Morales . The song was directed to be a dance @-@ club song , further broadening Carey 's " musical horizon . " The song incorporated Carey 's vocals , and added them to a thumping house beat , something he would do for many of her future singles . " Looking In " was the final song on the album . It was Carey 's most personal song at the time , one in which she let herself appear " naked " and " stripped down ; it was written by her and Afanasieff . " According to author Chris Nickson :
| [
{
"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 \" Melt Away \" was a song Carey produced on her own , and co @-@ wrote with Babyface . The song 's writing and production were \" superb . \" with each verse gliding into its chorus . According to Chris Nickson , \" Underneath the Stars \" was as \" strong as any slow jam released in the nineties , and one that would find a lot of flavor late at night with dancers . \" Another song that brought back reminders of older decades was \" Forever . \" The throwback was featured through the chord changes and in the way the guitar arpeggios \" stayed at the forefront of the music . \" The song displayed subtle vocals from Carey , as well as an undeniable richness . \" Daydream Interlude ( Sweet Fantasy Dub Mix ) \" was one of the liveliest tracks on the album . The song was a club remix of \" Fantasy \" , which was tuned and remixed by famed house music producer David Morales . The song was directed to be a dance @-@ club song , further broadening Carey 's \" musical horizon . \" The song incorporated Carey 's vocals , and added them to a thumping house beat , something he would do for many of her future singles . \" Looking In \" was the final song on the album . It was Carey 's most personal song at the time , one in which she let herself appear \" naked \" and \" stripped down ; it was written by her and Afanasieff . \" According to author Chris Nickson : \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 song ] reflected on her life now , the changes she 'd gone through , and the difference between the public perception of Mariah Carey and the real person . Intimate and revealing , it made an appropriate end to the album , and was evident that Mariah was growing , changing , and becoming much more herself , confident of who she was and what she could do . "
| [
{
"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 song ] reflected on her life now , the changes she 'd gone through , and the difference between the public perception of Mariah Carey and the real person . Intimate and revealing , it made an appropriate end to the album , and was evident that Mariah was growing , changing , and becoming much more herself , confident of who she was and what she could do . \" \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 song reflected on her life now, the changes she'd gone through, and <|end-PI|> <|retrieval|> <|start-KFI|> the difference between the public perception of Mariah Carey and the real person <|end-KFI|>. <|start-PI|> Intimate and revealing, it made an appropriate end to the album, and was evident that <|end-PI|> <|retrieval|> <|start-KFI|> Mariah was growing, changing, and becoming much more herself, confident of who she was and what she could do <|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of the song's reflection, while the KFI provides specific details about its content. The <|retrieval|> token is inserted to indicate where additional information about the song's themes might be needed.",
"Ideal Knowledge": "Detailed analysis of Mariah Carey's song lyrics, discussing themes of personal growth, public image vs. private reality, and the artist's self-reflection.",
"Key Factual Information": "the difference between the public perception of Mariah Carey and the real person",
"Prefix reformulated": "What did Mariah Carey's song reflect about her life?",
"Prefix-intent": "The song reflected on her life now, the changes she'd gone through, and",
"Verification": "The KFI directly provides information about what the song reflected, as prompted by the PI."
},
{
"Explanation": "The PI introduces the song's impact and sets up a query about what it revealed, while the KFI provides specific details about Mariah's personal growth. The <|retrieval|> token is inserted to indicate where additional information about Mariah's career development might be needed.",
"Ideal Knowledge": "Information on Mariah Carey's career progression, personal development, and artistic evolution, particularly focusing on the period surrounding this album's release.",
"Key Factual Information": "Mariah was growing, changing, and becoming much more herself, confident of who she was and what she could do",
"Prefix reformulated": "What was evident about Mariah Carey from this intimate and revealing song at the end of the album?",
"Prefix-intent": "Intimate and revealing, it made an appropriate end to the album, and was evident that",
"Verification": "The KFI directly describes Mariah Carey's personal growth and confidence, as prompted by the PI."
}
]
} |
Daydream received universal acclaim at the time of its release . Reviews applauded the little changes of style from previous Carey releases , some of whom adding that it is her best record , while others dismissed it as soulless and lacking of originality . AllMusic 's senior editor , Stephen Thomas Erlewine , awarded the album four and a half out of five stars . Erlewine called Daydream her " best record yet " and wrote , " Mariah Carey certainly knows how to construct an album . Positioning herself directly between urban R & B with tracks like " Fantasy , " and the adult contemporary radio format with songs like " One Sweet Day , " a duet with Boyz II Men , Carey appeals to both audiences equally because of the sheer amount of craft and hard work she puts into her albums . Daydream is her best record to date , featuring a consistently strong selection of songs and a remarkably impassioned performance by Carey . Daydream demonstrates that Carey continues to perfect her craft , and that she has earned her status as an R & B / pop diva . " In his review for the album , Ken Tucker from Entertainment Weekly called Daydream " her best record since her 1990 debut " , writing , " in fact , it 's easily the best collection Carey has put out since her self @-@ titled 1990 debut , the album Daydream most resembles in its emphasis on R & B grooves . " Tucker specifically complimented " One Sweet Day " , " Always Be My Baby " , " Forever " , and " Daydream Interlude " ( Fantasy Sweet Dub Mix ) , writing " One Sweet Day , her collaboration with Boyz II Men , radiates a breezy sexiness that Carey , for all the brazen hussiness of her public persona , rarely permits herself to reveal in song . I like the relaxed swing of " Always Be My Baby " , and the brisk waltz tempo of Forever . However , it 's on what many Carey fans will probably find the most throwaway cut , " Daydream Interlude ( Fantasy Sweet Dub Mix ) " , that the singer really defines herself . At her best , as she is on this clipped , spunky track , Carey is a disco diva for the ' 90s , a worthy successor to trailblazing women like Donna Summer and Vicki Sue Robinson , R & B singers with an affinity for the endless groove . "
| [
{
"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 Daydream received universal acclaim at the time of its release . Reviews applauded the little changes of style from previous Carey releases , some of whom adding that it is her best record , while others dismissed it as soulless and lacking of originality . AllMusic 's senior editor , Stephen Thomas Erlewine , awarded the album four and a half out of five stars . Erlewine called Daydream her \" best record yet \" and wrote , \" Mariah Carey certainly knows how to construct an album . Positioning herself directly between urban R & B with tracks like \" Fantasy , \" and the adult contemporary radio format with songs like \" One Sweet Day , \" a duet with Boyz II Men , Carey appeals to both audiences equally because of the sheer amount of craft and hard work she puts into her albums . Daydream is her best record to date , featuring a consistently strong selection of songs and a remarkably impassioned performance by Carey . Daydream demonstrates that Carey continues to perfect her craft , and that she has earned her status as an R & B / pop diva . \" In his review for the album , Ken Tucker from Entertainment Weekly called Daydream \" her best record since her 1990 debut \" , writing , \" in fact , it 's easily the best collection Carey has put out since her self @-@ titled 1990 debut , the album Daydream most resembles in its emphasis on R & B grooves . \" Tucker specifically complimented \" One Sweet Day \" , \" Always Be My Baby \" , \" Forever \" , and \" Daydream Interlude \" ( Fantasy Sweet Dub Mix ) , writing \" One Sweet Day , her collaboration with Boyz II Men , radiates a breezy sexiness that Carey , for all the brazen hussiness of her public persona , rarely permits herself to reveal in song . I like the relaxed swing of \" Always Be My Baby \" , and the brisk waltz tempo of Forever . However , it 's on what many Carey fans will probably find the most throwaway cut , \" Daydream Interlude ( Fantasy Sweet Dub Mix ) \" , that the singer really defines herself . At her best , as she is on this clipped , spunky track , Carey is a disco diva for the ' 90s , a worthy successor to trailblazing women like Donna Summer and Vicki Sue Robinson , R & B singers with an affinity for the endless groove . \" \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|>Daydream received universal acclaim at the time of its release . Reviews applauded<|end-PI|> <|retrieval|> <|start-KFI|>the little changes of style from previous Carey releases , some of whom adding that it is her best record , while others dismissed it as soulless and lacking of originality<|end-KFI|> . <|start-PI|>AllMusic 's senior editor , Stephen Thomas Erlewine , awarded the album<|end-PI|> <|retrieval|> <|start-KFI|>four and a half out of five stars<|end-KFI|> . Erlewine called Daydream her \" best record yet \" and wrote , \" Mariah Carey certainly knows how to construct an album . Positioning herself directly between urban R & B with tracks like \" Fantasy , \" and the adult contemporary radio format with songs like \" One Sweet Day , \" a duet with Boyz II Men , Carey appeals to both audiences equally because of the sheer amount of craft and hard work she puts into her albums . Daydream is her best record to date , featuring a consistently strong selection of songs and a remarkably impassioned performance by Carey . Daydream demonstrates that Carey continues to perfect her craft , and that she has earned her status as an R & B / pop diva . \" In his review for the album , <|start-PI|>Ken Tucker from Entertainment Weekly called Daydream<|end-PI|> <|retrieval|> <|start-KFI|>\" her best record since her 1990 debut \"<|end-KFI|> , writing , \" in fact , it 's easily the best collection Carey has put out since her self @-@ titled 1990 debut , the album Daydream most resembles in its emphasis on R & B grooves . \" Tucker specifically complimented \" One Sweet Day \" , \" Always Be My Baby \" , \" Forever \" , and \" Daydream Interlude \" ( Fantasy Sweet Dub Mix ) , writing \" One Sweet Day , her collaboration with Boyz II Men , radiates a breezy sexiness that Carey , for all the brazen hussiness of her public persona , rarely permits herself to reveal in song . I like the relaxed swing of \" Always Be My Baby \" , and the brisk waltz tempo of Forever . However , it 's on what many Carey fans will probably find the most throwaway cut , \" Daydream Interlude ( Fantasy Sweet Dub Mix ) \" , that the singer really defines herself . At her best , as she is on this clipped , spunky track , Carey is a disco diva for the ' 90s , a worthy successor to trailblazing women like Donna Summer and Vicki Sue Robinson , R & B singers with an affinity for the endless groove . \"",
"none": null,
"results": [
{
"Explanation": "The PI introduces the album's reception, and the KFI provides specific details about the reviews. The <|retrieval|> token is inserted to indicate where additional information about critical reception might be needed.",
"Ideal Knowledge": "A comprehensive summary of critical reviews for Mariah Carey's album Daydream, including comparisons to her previous works and contrasting opinions from different critics.",
"Key Factual Information": "the little changes of style from previous Carey releases , some of whom adding that it is her best record , while others dismissed it as soulless and lacking of originality",
"Prefix reformulated": "What did reviews applaud about Mariah Carey's album Daydream?",
"Prefix-intent": "Daydream received universal acclaim at the time of its release . Reviews applauded",
"Verification": "The KFI directly provides information about the reviews' reactions to the album, as prompted by the PI."
},
{
"Explanation": "The PI introduces a specific reviewer and sets up a query about his rating, while the KFI provides the exact rating. The <|retrieval|> token is inserted to indicate where additional information about the review might be needed.",
"Ideal Knowledge": "A detailed review of Mariah Carey's Daydream album by Stephen Thomas Erlewine, including his rating and justification for the score.",
"Key Factual Information": "four and a half out of five stars",
"Prefix reformulated": "What rating did AllMusic's senior editor Stephen Thomas Erlewine give to Mariah Carey's album Daydream?",
"Prefix-intent": "AllMusic 's senior editor , Stephen Thomas Erlewine , awarded the album",
"Verification": "The KFI directly provides the rating given by the editor, as prompted by the PI."
},
{
"Explanation": "The PI introduces another reviewer and his opinion, while the KFI provides a direct quote of his assessment. The <|retrieval|> token is inserted to indicate where additional information about this review might be needed.",
"Ideal Knowledge": "Ken Tucker's full review of Mariah Carey's Daydream album for Entertainment Weekly, including comparisons to her previous works and specific praise for the album.",
"Key Factual Information": "\" her best record since her 1990 debut \"",
"Prefix reformulated": "How did Ken Tucker from Entertainment Weekly describe Mariah Carey's album Daydream?",
"Prefix-intent": "Ken Tucker from Entertainment Weekly called Daydream",
"Verification": "The KFI directly quotes Tucker's description of the album, as prompted by the PI."
}
]
} |
Stephen Holden , editor of The New York Times , gave the album a positive review . Holden wrote the following regarding Daydream " Ms. Carey 's songwriting has taken a leap forward , becoming more relaxed , sexier , and less reliant on thudding cliches . " Holden praised " Fantasy " , which he wrote " with ' Fantasy , ' Ms. Carey glides confidently into the territory where gospel @-@ flavored pop @-@ soul meets light hip @-@ hop and recorded some of the most gorgeously spun choral music to be found on a contemporary album . " Additionally , he complimented " One Sweet Day , " " Melt Away , " " Always Be My Baby " , and " Underneath the Stars " , calling them " the best on the album " . People gave the album a positive review , calling it " her fourth and best album . " Additionally , People praised the album and its songs , writing " Daydream vaults over its pop predecessors because the material is both funkier and mellower . Carey also has better control of her instrument — her voice evincing greater muscularity and agility . She still pours it on a little thick at times when it comes to fervor , as on the midtempo ' Melt Away , ' which Carey co @-@ wrote with Babyface . For the most part she buzzes from strength to strength , from the bravura belting on ' One Sweet Day , ' a duet with Boyz II Men , to the rich gospel feel of ' I Am Free , ' which has a mood so churchy you can almost hear the ladies ' handheld fans snapping . " While the album was positively reviewed by critics , Carey 's cover of Journey 's 1982 song " Open Arms " was universally panned . Stephen Thomas Erlewine criticized the song , calling it " second rate " . " Open Arms " received a negative review from Stephen Holden as well , who called it a " sobbing remake " .
| [
{
"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 Stephen Holden , editor of The New York Times , gave the album a positive review . Holden wrote the following regarding Daydream \" Ms. Carey 's songwriting has taken a leap forward , becoming more relaxed , sexier , and less reliant on thudding cliches . \" Holden praised \" Fantasy \" , which he wrote \" with ' Fantasy , ' Ms. Carey glides confidently into the territory where gospel @-@ flavored pop @-@ soul meets light hip @-@ hop and recorded some of the most gorgeously spun choral music to be found on a contemporary album . \" Additionally , he complimented \" One Sweet Day , \" \" Melt Away , \" \" Always Be My Baby \" , and \" Underneath the Stars \" , calling them \" the best on the album \" . People gave the album a positive review , calling it \" her fourth and best album . \" Additionally , People praised the album and its songs , writing \" Daydream vaults over its pop predecessors because the material is both funkier and mellower . Carey also has better control of her instrument — her voice evincing greater muscularity and agility . She still pours it on a little thick at times when it comes to fervor , as on the midtempo ' Melt Away , ' which Carey co @-@ wrote with Babyface . For the most part she buzzes from strength to strength , from the bravura belting on ' One Sweet Day , ' a duet with Boyz II Men , to the rich gospel feel of ' I Am Free , ' which has a mood so churchy you can almost hear the ladies ' handheld fans snapping . \" While the album was positively reviewed by critics , Carey 's cover of Journey 's 1982 song \" Open Arms \" was universally panned . Stephen Thomas Erlewine criticized the song , calling it \" second rate \" . \" Open Arms \" received a negative review from Stephen Holden as well , who called it a \" sobbing remake \" . \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|>Stephen Holden , editor of The New York Times , gave the album<|end-PI|> <|retrieval|> <|start-KFI|>a positive review<|end-KFI|> . <|start-PI|>Holden wrote the following regarding Daydream<|end-PI|> <|retrieval|> <|start-KFI|>\" Ms. Carey 's songwriting has taken a leap forward , becoming more relaxed , sexier , and less reliant on thudding cliches . \"<|end-KFI|> <|start-PI|>Holden praised \" Fantasy \" , which he wrote<|end-PI|> <|retrieval|> <|start-KFI|>\" with ' Fantasy , ' Ms. Carey glides confidently into the territory where gospel @-@ flavored pop @-@ soul meets light hip @-@ hop and recorded some of the most gorgeously spun choral music to be found on a contemporary album . \"<|end-KFI|> Additionally , he complimented \" One Sweet Day , \" \" Melt Away , \" \" Always Be My Baby \" , and \" Underneath the Stars \" , calling them \" the best on the album \" . <|start-PI|>People gave the album<|end-PI|> <|retrieval|> <|start-KFI|>a positive review , calling it \" her fourth and best album . \"<|end-KFI|> Additionally , People praised the album and its songs , writing \" Daydream vaults over its pop predecessors because the material is both funkier and mellower . Carey also has better control of her instrument — her voice evincing greater muscularity and agility . She still pours it on a little thick at times when it comes to fervor , as on the midtempo ' Melt Away , ' which Carey co @-@ wrote with Babyface . For the most part she buzzes from strength to strength , from the bravura belting on ' One Sweet Day , ' a duet with Boyz II Men , to the rich gospel feel of ' I Am Free , ' which has a mood so churchy you can almost hear the ladies ' handheld fans snapping . \" <|start-PI|>While the album was positively reviewed by critics , Carey 's cover of Journey 's 1982 song \" Open Arms \" was<|end-PI|> <|retrieval|> <|start-KFI|>universally panned<|end-KFI|> . Stephen Thomas Erlewine criticized the song , calling it \" second rate \" . \" Open Arms \" received a negative review from Stephen Holden as well , who called it a \" sobbing remake \" .",
"none": null,
"results": [
{
"Explanation": "The PI introduces Stephen Holden and his role, setting up the context for his review. The KFI provides the nature of the review. The <|retrieval|> token is inserted to indicate where additional information about the review might be needed.",
"Ideal Knowledge": "A compilation of critical reviews for Mariah Carey's album 'Daydream', specifically highlighting Stephen Holden's review from The New York Times.",
"Key Factual Information": "a positive review",
"Prefix reformulated": "What kind of review did Stephen Holden, editor of The New York Times, give to Mariah Carey's album Daydream?",
"Prefix-intent": "Stephen Holden , editor of The New York Times , gave the album",
"Verification": "The KFI directly provides the type of review given by Stephen Holden, as prompted by the PI."
},
{
"Explanation": "The PI sets up a direct quote from Holden's review, and the KFI provides the specific content of that quote. The <|retrieval|> token is inserted to indicate where the exact quote begins.",
"Ideal Knowledge": "Detailed quotes and analysis from Stephen Holden's review of Mariah Carey's 'Daydream' album, focusing on her songwriting development.",
"Key Factual Information": "Ms. Carey 's songwriting has taken a leap forward , becoming more relaxed , sexier , and less reliant on thudding cliches",
"Prefix reformulated": "What did Stephen Holden write about Mariah Carey's album Daydream?",
"Prefix-intent": "Holden wrote the following regarding Daydream",
"Verification": "The KFI directly quotes Holden's review of the album, as prompted by the PI."
},
{
"Explanation": "The PI introduces Holden's praise for 'Fantasy', and the KFI provides the specific content of that praise. The <|retrieval|> token is inserted to indicate where the exact quote begins.",
"Ideal Knowledge": "Detailed analysis of Mariah Carey's song 'Fantasy' from music critics, particularly Stephen Holden's review, focusing on its musical style and vocal performance.",
"Key Factual Information": "with ' Fantasy , ' Ms. Carey glides confidently into the territory where gospel @-@ flavored pop @-@ soul meets light hip @-@ hop and recorded some of the most gorgeously spun choral music to be found on a contemporary album",
"Prefix reformulated": "What did Stephen Holden write about Mariah Carey's song 'Fantasy'?",
"Prefix-intent": "Holden praised \" Fantasy \" , which he wrote",
"Verification": "The KFI directly quotes Holden's praise for the song 'Fantasy', as prompted by the PI."
},
{
"Explanation": "The PI sets up People magazine's review, and the KFI provides the nature of the review and a direct quote. The <|retrieval|> token is inserted to indicate where the specific review details begin.",
"Ideal Knowledge": "A compilation of reviews for Mariah Carey's 'Daydream' album from various publications, specifically highlighting People magazine's review and rating.",
"Key Factual Information": "a positive review , calling it \" her fourth and best album \"",
"Prefix reformulated": "What kind of review did People magazine give to Mariah Carey's album Daydream?",
"Prefix-intent": "People gave the album",
"Verification": "The KFI directly provides the nature of People magazine's review and a quote from it, as prompted by the PI."
},
{
"Explanation": "The PI sets up the contrast between the overall positive album reviews and the reception of the 'Open Arms' cover, and the KFI provides the specific negative reception. The <|retrieval|> token is inserted to indicate where the critical consensus is stated.",
"Ideal Knowledge": "A compilation of critical reviews specifically for Mariah Carey's cover of Journey's 'Open Arms', highlighting the negative reception it received in contrast to the overall positive album reviews.",
"Key Factual Information": "universally panned",
"Prefix reformulated": "How was Mariah Carey's cover of Journey's song 'Open Arms' received by critics?",
"Prefix-intent": "While the album was positively reviewed by critics , Carey 's cover of Journey 's 1982 song \" Open Arms \" was",
"Verification": "The KFI directly states the critical reception of Carey's cover of 'Open Arms', as prompted by the PI."
}
]
} |
Daydream entered the Billboard 200 at number one , with 224 @,@ 000 copies sold , staying at the top spot the following week with 216 @,@ 000 copies sold , for a third consecutive week , it topped the charts with 170 @,@ 000 copies sold . It gained power again in the upcoming weeks of holiday sales where it peaked in the year 's last weeks with 486 @,@ 000 and 760 @,@ 000 units sold at the pinnacle . The album moved 760 @,@ 000 copies during the Christmas week of 1995 , the album 's highest sales week . It also reached number one on Top R & B / Hip @-@ Hop Albums chart . Daydream was the second best @-@ selling album of 1996 , and the eighteenth best @-@ selling album on the 1990s decade in the US . In the United States , Daydream became Carey 's best @-@ selling album , being certified diamond by the RIAA ( RIAA ) , denoting shipments of ten million copies .
| [
{
"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 Daydream entered the Billboard 200 at number one , with 224 @,@ 000 copies sold , staying at the top spot the following week with 216 @,@ 000 copies sold , for a third consecutive week , it topped the charts with 170 @,@ 000 copies sold . It gained power again in the upcoming weeks of holiday sales where it peaked in the year 's last weeks with 486 @,@ 000 and 760 @,@ 000 units sold at the pinnacle . The album moved 760 @,@ 000 copies during the Christmas week of 1995 , the album 's highest sales week . It also reached number one on Top R & B / Hip @-@ Hop Albums chart . Daydream was the second best @-@ selling album of 1996 , and the eighteenth best @-@ selling album on the 1990s decade in the US . In the United States , Daydream became Carey 's best @-@ selling album , being certified diamond by the RIAA ( RIAA ) , denoting shipments of ten million copies . \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 Canada , Daydream peaked at number two on the charts , and was certified seven @-@ times platinum by the Canadian Recording Industry Association ( CRIA ) . The album experienced success in Europe , where it reached number one in Germany , The Netherlands , Switzerland and the United Kingdom . In France , Daydream peaked at number two and was certified double @-@ platinum by the Syndicat National de l 'Édition Phonographique ( SNEP ) . Sales in France are estimated at 800 @,@ 000 copies . Daydream was certified triple @-@ platinum by the International Federation of the Phonographic Industry ( IFPI ) , denoting shipments of three million copies throughout Europe in 1996 .
| [
{
"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 Canada , Daydream peaked at number two on the charts , and was certified seven @-@ times platinum by the Canadian Recording Industry Association ( CRIA ) . The album experienced success in Europe , where it reached number one in Germany , The Netherlands , Switzerland and the United Kingdom . In France , Daydream peaked at number two and was certified double @-@ platinum by the Syndicat National de l 'Édition Phonographique ( SNEP ) . Sales in France are estimated at 800 @,@ 000 copies . Daydream was certified triple @-@ platinum by the International Federation of the Phonographic Industry ( IFPI ) , denoting shipments of three million copies throughout Europe in 1996 . \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 Canada, Daydream peaked at number<|end-PI|> <|retrieval|> <|start-KFI|>two on the charts, and was certified seven-times platinum by the Canadian Recording Industry Association (CRIA)<|end-KFI|>. <|start-PI|>The album experienced success in Europe, where it reached number one in<|end-PI|> <|retrieval|> <|start-KFI|>Germany, The Netherlands, Switzerland and the United Kingdom<|end-KFI|>. <|start-PI|>In France, Daydream peaked at number two and was certified<|end-PI|> <|retrieval|> <|start-KFI|>double-platinum by the Syndicat National de l'Édition Phonographique (SNEP)<|end-KFI|>. <|start-PI|>Sales in France are estimated at<|end-PI|> <|retrieval|> <|start-KFI|>800,000 copies<|end-KFI|>. <|start-PI|>Daydream was certified triple-platinum by the International Federation of the Phonographic Industry (IFPI), denoting shipments of<|end-PI|> <|retrieval|> <|start-KFI|>three million copies throughout Europe in 1996<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces the album's performance in Canada, while the KFI provides specific chart position and certification details. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "A comprehensive list of chart positions and certifications for the album 'Daydream' in various countries, with specific focus on its performance in Canada.",
"Key Factual Information": "two on the charts, and was certified seven-times platinum by the Canadian Recording Industry Association (CRIA)",
"Prefix reformulated": "What position did Daydream peak at on the charts in Canada?",
"Prefix-intent": "In Canada, Daydream peaked at number",
"Verification": "The KFI directly provides the chart position and certification information requested in the PI."
},
{
"Explanation": "The PI sets up a query about the album's success in Europe, and the KFI provides the specific countries where it reached number one. The <|retrieval|> token is inserted to indicate where additional information about European chart performance might be needed.",
"Ideal Knowledge": "A detailed breakdown of 'Daydream's' chart performance across European countries, highlighting those where it reached the top position.",
"Key Factual Information": "Germany, The Netherlands, Switzerland and the United Kingdom",
"Prefix reformulated": "In which European countries did the album 'Daydream' reach number one?",
"Prefix-intent": "The album experienced success in Europe, where it reached number one in",
"Verification": "The KFI directly lists the European countries where the album reached number one, as prompted by the PI."
},
{
"Explanation": "The PI introduces the album's performance in France, and the KFI provides the specific certification details. The <|retrieval|> token is inserted to indicate where additional information about French music certifications might be needed.",
"Ideal Knowledge": "Detailed information about music certifications in France, including the criteria for double-platinum certification by SNEP and its significance in the French music market.",
"Key Factual Information": "double-platinum by the Syndicat National de l'Édition Phonographique (SNEP)",
"Prefix reformulated": "What certification did the album 'Daydream' receive in France?",
"Prefix-intent": "In France, Daydream peaked at number two and was certified",
"Verification": "The KFI directly provides the certification information for France, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the album's sales in France, and the KFI provides the specific estimated figure. The <|retrieval|> token is inserted to indicate where additional information about album sales might be needed.",
"Ideal Knowledge": "Detailed sales data for 'Daydream' in various countries, with specific focus on its performance in the French market and how it compares to other markets.",
"Key Factual Information": "800,000 copies",
"Prefix reformulated": "What are the estimated sales of 'Daydream' in France?",
"Prefix-intent": "Sales in France are estimated at",
"Verification": "The KFI directly provides the estimated sales figure for France, as prompted by the PI."
},
{
"Explanation": "The PI introduces the IFPI certification, and the KFI provides the specific number of copies shipped and the year. The <|retrieval|> token is inserted to indicate where additional information about IFPI certifications and European music market performance might be needed.",
"Ideal Knowledge": "Information on IFPI certifications, including the criteria for triple-platinum status in Europe, and how it relates to actual sales and shipments across the continent.",
"Key Factual Information": "three million copies throughout Europe in 1996",
"Prefix reformulated": "How many copies of 'Daydream' were shipped throughout Europe according to IFPI certification?",
"Prefix-intent": "Daydream was certified triple-platinum by the International Federation of the Phonographic Industry (IFPI), denoting shipments of",
"Verification": "The KFI directly provides the number of copies shipped in Europe and the year, as prompted by the PI."
}
]
} |
In Australia , Daydream was certified five @-@ times platinum by the Australian Recording Industry Association ( ARIA ) , denoting shipments of 350 @,@ 000 copies . The album finished ninth on the ARIA End of Year Charts in both 1995 and 1996 . In Japan , the album debuted at number one on the Oricon charts . According to the Oricon , Daydream made the top five of the best @-@ selling albums in Japan by a non @-@ Asian artist , with 2 @.@ 5 million copies sold . Daydream remains one of the best @-@ selling albums of all time , with sales of 25 million copies 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 In Australia , Daydream was certified five @-@ times platinum by the Australian Recording Industry Association ( ARIA ) , denoting shipments of 350 @,@ 000 copies . The album finished ninth on the ARIA End of Year Charts in both 1995 and 1996 . In Japan , the album debuted at number one on the Oricon charts . According to the Oricon , Daydream made the top five of the best @-@ selling albums in Japan by a non @-@ Asian artist , with 2 @.@ 5 million copies sold . Daydream remains one of the best @-@ selling albums of all time , with sales of 25 million copies 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
} |
Six singles were released from Daydream . " Fantasy " was released as the album 's lead single on September 12 , 1995 . The song debuted at number one on the Billboard Hot 100 , making Carey the first female artist , and the second performer ever to accomplish the feat . The song spent eight consecutive weeks atop the chart , as well as topping the charts in Australia , Canada and New Zealand . In Europe , " Fantasy " performed well , peaking within the top five in Belgium , Finland , France , and the United Kingdom . " Fantasy " was the second best @-@ selling single of 1995 in the US , with sales of 1 @.@ 5 million in 95 ' alone . " One Sweet Day " was chosen as the follow @-@ up single , achieving similar success . The song once again debuted at the top of the US charts , and became the longest running number one single in US history , spending sixteen consecutive weeks atop the Billboard charts . The song became a success in other regions around the world as well , topping the charts in Canada and New Zealand and peaked within the top five in Australia , France , Ireland , and The Netherlands . Serving as the album 's third single in select European countries , " Open Arms " was released on December 5 , 1995 . The song achieved success in the UK , Ireland and New Zealand , where it peaked at number four , seven and eight . However , " Open Arms " charted weakly in other European countries , such as Belgium , France , and Germany , where it charted outside the top @-@ thirty .
| [
{
"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 Six singles were released from Daydream . \" Fantasy \" was released as the album 's lead single on September 12 , 1995 . The song debuted at number one on the Billboard Hot 100 , making Carey the first female artist , and the second performer ever to accomplish the feat . The song spent eight consecutive weeks atop the chart , as well as topping the charts in Australia , Canada and New Zealand . In Europe , \" Fantasy \" performed well , peaking within the top five in Belgium , Finland , France , and the United Kingdom . \" Fantasy \" was the second best @-@ selling single of 1995 in the US , with sales of 1 @.@ 5 million in 95 ' alone . \" One Sweet Day \" was chosen as the follow @-@ up single , achieving similar success . The song once again debuted at the top of the US charts , and became the longest running number one single in US history , spending sixteen consecutive weeks atop the Billboard charts . The song became a success in other regions around the world as well , topping the charts in Canada and New Zealand and peaked within the top five in Australia , France , Ireland , and The Netherlands . Serving as the album 's third single in select European countries , \" Open Arms \" was released on December 5 , 1995 . The song achieved success in the UK , Ireland and New Zealand , where it peaked at number four , seven and eight . However , \" Open Arms \" charted weakly in other European countries , such as Belgium , France , and Germany , where it charted outside the top @-@ thirty . \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
} |
" Always Be My Baby " was released as the fourth single . The song debuted at number two on the Hot 100 , failing to become Carey 's third number one debut ( a feat she would accomplish with " Honey " in 1997 ) . The song eventually reached and stayed atop the charts for two weeks , and then kept steady at number two for nine weeks . " Always Be My Baby " performed moderately in other major markets . The song peaked within the top five in Canada , New Zealand and the UK , but charted outside the top @-@ ten elsewhere . " Forever " was chosen as the fifth single from Daydream . The song was ineligible to chart in the Hot 100 , but managed to crack the top @-@ ten on the Hot 100 airplay chart . " Forever " charted well in Canada , where it peaked at number thirteen . " Underneath the Stars " was chosen as the sixth and final single from the album . Described by Carey as one of her favorite songs , " Underneath the Stars " received a limited number of pressings in the US , where it charted weakly on the Billboard Hot R & B / Hip @-@ Hop Songs .
| [
{
"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 \" Always Be My Baby \" was released as the fourth single . The song debuted at number two on the Hot 100 , failing to become Carey 's third number one debut ( a feat she would accomplish with \" Honey \" in 1997 ) . The song eventually reached and stayed atop the charts for two weeks , and then kept steady at number two for nine weeks . \" Always Be My Baby \" performed moderately in other major markets . The song peaked within the top five in Canada , New Zealand and the UK , but charted outside the top @-@ ten elsewhere . \" Forever \" was chosen as the fifth single from Daydream . The song was ineligible to chart in the Hot 100 , but managed to crack the top @-@ ten on the Hot 100 airplay chart . \" Forever \" charted well in Canada , where it peaked at number thirteen . \" Underneath the Stars \" was chosen as the sixth and final single from the album . Described by Carey as one of her favorite songs , \" Underneath the Stars \" received a limited number of pressings in the US , where it charted weakly on the Billboard Hot R & B / Hip @-@ Hop Songs . \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|> \"Always Be My Baby\" was released as the fourth single . The song debuted at number <|end-PI|> <|retrieval|> <|start-KFI|> two on the Hot 100 , failing to become Carey 's third number one debut ( a feat she would accomplish with \" Honey \" in 1997 ) <|end-KFI|> . <|start-PI|> The song eventually reached and stayed atop the charts for <|end-PI|> <|retrieval|> <|start-KFI|> two weeks , and then kept steady at number two for nine weeks <|end-KFI|> . <|start-PI|> \"Always Be My Baby\" performed moderately in other major markets . The song peaked <|end-PI|> <|retrieval|> <|start-KFI|> within the top five in Canada , New Zealand and the UK , but charted outside the top @-@ ten elsewhere <|end-KFI|> . <|start-PI|> \"Forever\" was chosen as the fifth single from Daydream . The song was ineligible to chart in the Hot 100 , but managed to <|end-PI|> <|retrieval|> <|start-KFI|> crack the top @-@ ten on the Hot 100 airplay chart <|end-KFI|> . <|start-PI|> \"Forever\" charted well in Canada , where it peaked at number <|end-PI|> <|retrieval|> <|start-KFI|> thirteen <|end-KFI|> . <|start-PI|> \"Underneath the Stars\" was chosen as the sixth and final single from the album . Described by Carey as one of her favorite songs , \"Underneath the Stars\" received <|end-PI|> <|retrieval|> <|start-KFI|> a limited number of pressings in the US , where it charted weakly on the Billboard Hot R & B / Hip @-@ Hop Songs <|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the song and sets up a query about its chart performance, while the KFI provides the specific debut position and additional context. The <|retrieval|> token is inserted to indicate where additional chart information might be needed.",
"Ideal Knowledge": "A detailed chart history of Mariah Carey's singles, specifically focusing on their debut positions on the Hot 100 chart, including information about her number one debuts.",
"Key Factual Information": "two on the Hot 100 , failing to become Carey 's third number one debut ( a feat she would accomplish with \" Honey \" in 1997 )",
"Prefix reformulated": "What was the debut position of Mariah Carey's \"Always Be My Baby\" on the Hot 100 chart?",
"Prefix-intent": "\"Always Be My Baby\" was released as the fourth single . The song debuted at number",
"Verification": "The KFI directly provides the debut position of the song on the Hot 100 chart, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the song's chart longevity, and the KFI provides specific information about its time at number one and number two. The <|retrieval|> token is inserted to indicate where additional chart performance data might be needed.",
"Ideal Knowledge": "Detailed chart performance data for \"Always Be My Baby,\" including its week-by-week positions and duration at various chart positions.",
"Key Factual Information": "two weeks , and then kept steady at number two for nine weeks",
"Prefix reformulated": "How long did Mariah Carey's \"Always Be My Baby\" stay at number one on the charts?",
"Prefix-intent": "The song eventually reached and stayed atop the charts for",
"Verification": "The KFI directly provides the duration of the song's stay at number one and its subsequent position, as prompted by the PI."
},
{
"Explanation": "The PI introduces the song's international performance, and the KFI provides specific information about its chart positions in different countries. The <|retrieval|> token is inserted to indicate where additional international chart data might be needed.",
"Ideal Knowledge": "A comprehensive list of international chart positions for \"Always Be My Baby,\" including specific peak positions in various countries.",
"Key Factual Information": "within the top five in Canada , New Zealand and the UK , but charted outside the top @-@ ten elsewhere",
"Prefix reformulated": "How did Mariah Carey's \"Always Be My Baby\" perform in international markets?",
"Prefix-intent": "\"Always Be My Baby\" performed moderately in other major markets . The song peaked",
"Verification": "The KFI directly provides information about the song's performance in various international markets, as prompted by the PI."
},
{
"Explanation": "The PI introduces the song and its chart ineligibility, setting up a query about its performance, while the KFI provides specific information about its success on an alternative chart. The <|retrieval|> token is inserted to indicate where additional chart information might be needed.",
"Ideal Knowledge": "Detailed information about Billboard chart rules and alternative charts, specifically focusing on the Hot 100 airplay chart and how songs ineligible for the main Hot 100 can still achieve success.",
"Key Factual Information": "crack the top @-@ ten on the Hot 100 airplay chart",
"Prefix reformulated": "How did Mariah Carey's \"Forever\" perform on the charts despite being ineligible for the Hot 100?",
"Prefix-intent": "\"Forever\" was chosen as the fifth single from Daydream . The song was ineligible to chart in the Hot 100 , but managed to",
"Verification": "The KFI directly provides information about the song's performance on an alternative chart, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the song's performance in Canada, and the KFI provides the specific peak position. The <|retrieval|> token is inserted to indicate where additional chart data for Canada might be needed.",
"Ideal Knowledge": "A comprehensive list of international chart positions for \"Forever,\" with a focus on its performance in Canada, including its peak position and duration on the chart.",
"Key Factual Information": "thirteen",
"Prefix reformulated": "What was the peak chart position of Mariah Carey's \"Forever\" in Canada?",
"Prefix-intent": "\"Forever\" charted well in Canada , where it peaked at number",
"Verification": "The KFI directly provides the peak chart position in Canada, as prompted by the PI."
},
{
"Explanation": "The PI introduces the song and sets up a query about its release and reception, while the KFI provides specific information about its limited release and chart performance. The <|retrieval|> token is inserted to indicate where additional information about the song's release and chart performance might be needed.",
"Ideal Knowledge": "Detailed information about the release strategy for \"Underneath the Stars,\" including the number of pressings, promotional efforts, and its performance on various Billboard charts, particularly the Hot R&B/Hip-Hop Songs chart.",
"Key Factual Information": "a limited number of pressings in the US , where it charted weakly on the Billboard Hot R & B / Hip @-@ Hop Songs",
"Prefix reformulated": "How was Mariah Carey's \"Underneath the Stars\" released and received in the US?",
"Prefix-intent": "\"Underneath the Stars\" was chosen as the sixth and final single from the album . Described by Carey as one of her favorite songs , \"Underneath the Stars\" received",
"Verification": "The KFI directly provides information about the song's release and chart performance in the US, as prompted by the PI."
}
]
} |
In order to promote the album , Carey embarked on her second head @-@ lining tour . Originally , she had not planned to tour , due to the long travel times and hassle ; however , after many requests from fans , Carey agreed to tour . The tour reached Japan and select European countries , not visiting the United States . This was possibly due to the mixed reception Carey 's 1993 North American Music Box Tour received three years prior . The shows were all spaced apart , giving Carey time to rest her vocals , " It 's very strenuous to sing all my songs back to back , but I 'm actually really looking forward to it . " Many musicians joined Carey for the tour , including Randy Jackson who served as the musical director and played the bass , Dan Shea on the keyboards , Vernon Black playing the guitar , Gigi Conway on the drums , and percussion and music sequencing by Peter Michael and Gary Cirimelli . All of the musicians and background vocalists were under the supervision of Walter Afanasieff , who played the piano and guided the production . Before embarking on her world tour in 1996 , Carey performed a sold @-@ out show at Madison Square Garden in 1995 . The performance was filmed , and released as a DVD titled Fantasy : Mariah Carey at Madison Square Garden . It became Carey 's fourth video release .
| [
{
"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 order to promote the album , Carey embarked on her second head @-@ lining tour . Originally , she had not planned to tour , due to the long travel times and hassle ; however , after many requests from fans , Carey agreed to tour . The tour reached Japan and select European countries , not visiting the United States . This was possibly due to the mixed reception Carey 's 1993 North American Music Box Tour received three years prior . The shows were all spaced apart , giving Carey time to rest her vocals , \" It 's very strenuous to sing all my songs back to back , but I 'm actually really looking forward to it . \" Many musicians joined Carey for the tour , including Randy Jackson who served as the musical director and played the bass , Dan Shea on the keyboards , Vernon Black playing the guitar , Gigi Conway on the drums , and percussion and music sequencing by Peter Michael and Gary Cirimelli . All of the musicians and background vocalists were under the supervision of Walter Afanasieff , who played the piano and guided the production . Before embarking on her world tour in 1996 , Carey performed a sold @-@ out show at Madison Square Garden in 1995 . The performance was filmed , and released as a DVD titled Fantasy : Mariah Carey at Madison Square Garden . It became Carey 's fourth video release . \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
} |
When the three Japanese shows at the Tokyo Dome went on sale , Carey set a record after all 150 @,@ 000 tickets sold out under three hours . The shows became the fastest sellout in the stadium 's history , breaking the previous record held by The Rolling Stones . The shows in Japan were a critical and commercial success , with critics and fans raving about the show and Carey 's vocals . Carey 's presence in Asia in the 90s was unparalleled to any other international artist . Her international success and anticipation was even compared to the " Beatlemania " in the 1960s . In an interview with MTV , Carey spoke of how she felt performing in Asia :
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n When the three Japanese shows at the Tokyo Dome went on sale , Carey set a record after all 150 @,@ 000 tickets sold out under three hours . The shows became the fastest sellout in the stadium 's history , breaking the previous record held by The Rolling Stones . The shows in Japan were a critical and commercial success , with critics and fans raving about the show and Carey 's vocals . Carey 's presence in Asia in the 90s was unparalleled to any other international artist . Her international success and anticipation was even compared to the \" Beatlemania \" in the 1960s . In an interview with MTV , Carey spoke of how she felt performing in Asia : \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
" First of all , you 're in front of so many people that basically don 't speak your language . It took a little getting used to , but I think by the end of the show , you know , everybody started to kind of relax . "
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n \" First of all , you 're in front of so many people that basically don 't speak your language . It took a little getting used to , but I think by the end of the show , you know , everybody started to kind of relax . \" \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
} |
For the show , Carey sang fourteen original songs , including many of her biggest hits up until that point , as well as many songs from Daydream . They included " Fantasy , " " One Sweet Day , " " Open Arms , " " Always Be My Baby , " " Forever " and " Underneath the Stars " as well as hits from her previous studio efforts . Carey 's following shows in France , Germany , The Netherlands and the United Kingdom were all sold @-@ out as well , receiving warm critical response . During the span of the tour 's seven short dates , Carey had already begun working on concepts for her new album Butterfly . According to author Marc Shapiro , Carey 's European tour was truly a success , in many aspects :
| [
{
"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 For the show , Carey sang fourteen original songs , including many of her biggest hits up until that point , as well as many songs from Daydream . They included \" Fantasy , \" \" One Sweet Day , \" \" Open Arms , \" \" Always Be My Baby , \" \" Forever \" and \" Underneath the Stars \" as well as hits from her previous studio efforts . Carey 's following shows in France , Germany , The Netherlands and the United Kingdom were all sold @-@ out as well , receiving warm critical response . During the span of the tour 's seven short dates , Carey had already begun working on concepts for her new album Butterfly . According to author Marc Shapiro , Carey 's European tour was truly a success , in many aspects : \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": "For the show, <|start-PI|>Carey sang fourteen original songs, including<|end-PI|> <|retrieval|> <|start-KFI|>many of her biggest hits up until that point, as well as many songs from Daydream<|end-KFI|>. <|start-PI|>They included<|end-PI|> <|retrieval|> <|start-KFI|>\"Fantasy,\" \"One Sweet Day,\" \"Open Arms,\" \"Always Be My Baby,\" \"Forever\" and \"Underneath the Stars\" as well as hits from her previous studio efforts<|end-KFI|>. <|start-PI|>Carey's following shows in France, Germany, The Netherlands and the United Kingdom were<|end-PI|> <|retrieval|> <|start-KFI|>all sold-out as well, receiving warm critical response<|end-KFI|>. <|start-PI|>During the span of the tour's seven short dates, Carey had already begun working on<|end-PI|> <|retrieval|> <|start-KFI|>concepts for her new album Butterfly<|end-KFI|>. <|start-PI|>According to author Marc Shapiro, Carey's European tour was truly a success, in many aspects:<|end-PI|> <|retrieval|>",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of Carey's performance, while the KFI provides details about the song selection. The <|retrieval|> token is inserted to indicate where additional information about Carey's hit songs might be needed.",
"Ideal Knowledge": "A comprehensive list of Mariah Carey's biggest hits and songs from her album Daydream, along with their release dates and chart performances.",
"Key Factual Information": "many of her biggest hits up until that point, as well as many songs from Daydream",
"Prefix reformulated": "What songs did Mariah Carey sing for the show?",
"Prefix-intent": "For the show, Carey sang fourteen original songs, including",
"Verification": "The KFI directly provides information about the songs Carey sang, as requested in the PI."
},
{
"Explanation": "The PI introduces a list of songs, and the KFI provides the specific titles. The <|retrieval|> token is inserted to indicate where additional information about these songs might be needed.",
"Ideal Knowledge": "Detailed information about each of the listed songs, including their album origins, release dates, chart performances, and significance in Mariah Carey's career.",
"Key Factual Information": "\"Fantasy,\" \"One Sweet Day,\" \"Open Arms,\" \"Always Be My Baby,\" \"Forever\" and \"Underneath the Stars\" as well as hits from her previous studio efforts",
"Prefix reformulated": "What specific songs did Mariah Carey perform in the show?",
"Prefix-intent": "They included",
"Verification": "The KFI directly lists the specific songs performed, as prompted by the PI."
},
{
"Explanation": "The PI lists the European countries where Carey performed, and the KFI provides information about the shows' success. The <|retrieval|> token is inserted to indicate where additional information about the European tour's reception might be needed.",
"Ideal Knowledge": "Detailed information about Mariah Carey's European tour dates, ticket sales, venue capacities, and critical reviews from each country mentioned.",
"Key Factual Information": "all sold-out as well, receiving warm critical response",
"Prefix reformulated": "How were Mariah Carey's shows in Europe received?",
"Prefix-intent": "Carey's following shows in France, Germany, The Netherlands and the United Kingdom were",
"Verification": "The KFI directly provides information about the reception of Carey's European shows, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of Carey's activities during the tour, and the KFI provides specific information about her new album project. The <|retrieval|> token is inserted to indicate where additional information about the Butterfly album might be needed.",
"Ideal Knowledge": "Information about the development process of Mariah Carey's album Butterfly, including its concept, timeline, and how it was influenced by her experiences during the tour.",
"Key Factual Information": "concepts for her new album Butterfly",
"Prefix reformulated": "What was Mariah Carey working on during her tour?",
"Prefix-intent": "During the span of the tour's seven short dates, Carey had already begun working on",
"Verification": "The KFI directly provides information about Carey's work during the tour, as prompted by the PI."
},
{
"Explanation": "The PI introduces Marc Shapiro's assessment of the tour, but the passage doesn't provide specific details about his description. The <|retrieval|> token is inserted to indicate where additional information about Shapiro's analysis might be needed.",
"Ideal Knowledge": "Quotes or paraphrases from Marc Shapiro's work detailing the various aspects of success in Mariah Carey's European tour, such as ticket sales, critical reception, and impact on her career.",
"Key Factual Information": "",
"Prefix reformulated": "How does author Marc Shapiro describe the success of Mariah Carey's European tour?",
"Prefix-intent": "According to author Marc Shapiro, Carey's European tour was truly a success, in many aspects:",
"Verification": "The KFI is empty as there is no specific information provided about Marc Shapiro's description of the tour's success."
}
]
} |
" The European tour mirrored the success of her shows in Japan . Mariah 's appearances overseas were heartening experiences for the singer . The popularity – indeed , mania – surrounding her shows in these countries reached massive proportions . The reception the concerts received reflected the fact that Mariah 's music cut through race and language barriers and had struck at a universal , emotional core with fans . The tour cemented the fact that Mariah Carey had arrived as the performing centerpiece on the world stage . "
| [
{
"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 European tour mirrored the success of her shows in Japan . Mariah 's appearances overseas were heartening experiences for the singer . The popularity – indeed , mania – surrounding her shows in these countries reached massive proportions . The reception the concerts received reflected the fact that Mariah 's music cut through race and language barriers and had struck at a universal , emotional core with fans . The tour cemented the fact that Mariah Carey had arrived as the performing centerpiece on the world stage . \" \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 addition to touring the world , Carey performed on a variety of television programs and award shows . After " Fantasy " was released in September throughout Europe , Carey performed the song on the popular British chart show Top of the Pops , which aired live via satellite on Asian television . Carey performed " Fantasy " in France and at the 23rd Annual American Music Awards on January 29 , 1996 . " One Sweet Day " was performed at the 38th Annual Grammy Awards , Princess Diana 's memorial service in September 1997 , and at Carey 's Black Entertainment Television Christmas special in 2001 . During her European promotional tour for the album , Carey performed " Open Arms " on various television programs , including Wetten , dass .. ? in Germany , Top of the Pops and Des O 'Connor in the United Kingdom and on Swedish television .
| [
{
"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 addition to touring the world , Carey performed on a variety of television programs and award shows . After \" Fantasy \" was released in September throughout Europe , Carey performed the song on the popular British chart show Top of the Pops , which aired live via satellite on Asian television . Carey performed \" Fantasy \" in France and at the 23rd Annual American Music Awards on January 29 , 1996 . \" One Sweet Day \" was performed at the 38th Annual Grammy Awards , Princess Diana 's memorial service in September 1997 , and at Carey 's Black Entertainment Television Christmas special in 2001 . During her European promotional tour for the album , Carey performed \" Open Arms \" on various television programs , including Wetten , dass .. ? in Germany , Top of the Pops and Des O 'Connor in the United Kingdom and on Swedish television . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "In addition to touring the world, Carey performed on a variety of television programs and award shows. <|start-PI|>After \"Fantasy\" was released in September throughout Europe, Carey performed the song on<|end-PI|> <|retrieval|> <|start-KFI|>the popular British chart show Top of the Pops, which aired live via satellite on Asian television<|end-KFI|>. <|start-PI|>Carey performed \"Fantasy\" in France and at<|end-PI|> <|retrieval|> <|start-KFI|>the 23rd Annual American Music Awards on January 29, 1996<|end-KFI|>. <|start-PI|>\"One Sweet Day\" was performed at the 38th Annual Grammy Awards, Princess Diana's memorial service in September 1997, and at<|end-PI|> <|retrieval|> <|start-KFI|>Carey's Black Entertainment Television Christmas special in 2001<|end-KFI|>. <|start-PI|>During her European promotional tour for the album, Carey performed \"Open Arms\" on various television programs, including<|end-PI|> <|retrieval|> <|start-KFI|>Wetten, dass..? in Germany, Top of the Pops and Des O'Connor in the United Kingdom and on Swedish television<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of the song's release and introduces a performance, while the KFI specifies the exact show and its broadcast details. The <|retrieval|> token is inserted to indicate where additional information about the performance venue might be needed.",
"Ideal Knowledge": "A list of Mariah Carey's performances of \"Fantasy\" in 1995, including details about the Top of the Pops show, its popularity in Britain, and its broadcast reach in Asia.",
"Key Factual Information": "the popular British chart show Top of the Pops, which aired live via satellite on Asian television",
"Prefix reformulated": "Where did Mariah Carey perform \"Fantasy\" after its release in Europe?",
"Prefix-intent": "After \"Fantasy\" was released in September throughout Europe, Carey performed the song on",
"Verification": "The KFI directly provides the information about where Carey performed the song, as prompted by the PI."
},
{
"Explanation": "The PI mentions a performance in France and sets up for another performance location, while the KFI provides the specific award show and date. The <|retrieval|> token is inserted to indicate where additional information about the performance venues might be needed.",
"Ideal Knowledge": "A detailed list of Mariah Carey's performances of \"Fantasy\" in 1995-1996, including international performances and major award shows, with specific dates and locations.",
"Key Factual Information": "the 23rd Annual American Music Awards on January 29, 1996",
"Prefix reformulated": "At which event did Mariah Carey perform \"Fantasy\" in addition to France?",
"Prefix-intent": "Carey performed \"Fantasy\" in France and at",
"Verification": "The KFI directly provides the specific event and date where Carey performed \"Fantasy\", as prompted by the PI."
},
{
"Explanation": "The PI lists two performances of \"One Sweet Day\" and sets up for a third, while the KFI provides the specific television special and year. The <|retrieval|> token is inserted to indicate where additional information about this performance might be needed.",
"Ideal Knowledge": "A comprehensive list of Mariah Carey's performances of \"One Sweet Day\" from 1995 to 2001, including details about each performance, the significance of the events, and the song's popularity over time.",
"Key Factual Information": "Carey's Black Entertainment Television Christmas special in 2001",
"Prefix reformulated": "Where else did Mariah Carey perform \"One Sweet Day\" besides the Grammy Awards and Princess Diana's memorial?",
"Prefix-intent": "\"One Sweet Day\" was performed at the 38th Annual Grammy Awards, Princess Diana's memorial service in September 1997, and at",
"Verification": "The KFI directly provides an additional venue where Carey performed \"One Sweet Day\", as prompted by the PI."
},
{
"Explanation": "The PI introduces the context of the European tour and sets up for a list of television programs, while the KFI provides the specific shows and their countries. The <|retrieval|> token is inserted to indicate where additional information about these television appearances might be needed.",
"Ideal Knowledge": "A detailed schedule of Mariah Carey's European promotional tour for the album, including dates, locations, and specific television programs where she performed \"Open Arms\", along with information about the popularity and reach of each show.",
"Key Factual Information": "Wetten, dass..? in Germany, Top of the Pops and Des O'Connor in the United Kingdom and on Swedish television",
"Prefix reformulated": "On which television programs did Mariah Carey perform \"Open Arms\" during her European promotional tour?",
"Prefix-intent": "During her European promotional tour for the album, Carey performed \"Open Arms\" on various television programs, including",
"Verification": "The KFI directly lists the television programs where Carey performed \"Open Arms\", as prompted by the PI."
}
]
} |
The music industry took note of Carey 's success . She won two awards at the 1996 American Music Awards for her solo efforts : Favorite Pop / Rock Female Artist and Favorite Soul / R & B Female Artist . Throughout 1995 & 1996 , Carey was awarded various prestigious awards at the World Music Awards , including " World 's Best Selling Female R & B Artist " , " World 's Best Selling Overall Female Recording Artist , " " World 's Best Selling Pop Artist " and " World 's Best Selling Overall Recording Artist . " Additionally , " Fantasy " was named " Song of the Year " at the BMI Awards and " Favorite Song " at the Blockbuster Entertainment Awards , where Carey also won the award for " Top Pop Female . " In 1996 , Carey won many awards at the Billboard Music Awards , including " Hot 100 Singles Artist of the Year " , " Hot 100 Airplay ( Always Be My Baby ) , " " Hot Adult Contemporary Artist of the Year " and " Special Award for 16 weeks at # 1 for ' One Sweet Day . ' "
| [
{
"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 music industry took note of Carey 's success . She won two awards at the 1996 American Music Awards for her solo efforts : Favorite Pop / Rock Female Artist and Favorite Soul / R & B Female Artist . Throughout 1995 & 1996 , Carey was awarded various prestigious awards at the World Music Awards , including \" World 's Best Selling Female R & B Artist \" , \" World 's Best Selling Overall Female Recording Artist , \" \" World 's Best Selling Pop Artist \" and \" World 's Best Selling Overall Recording Artist . \" Additionally , \" Fantasy \" was named \" Song of the Year \" at the BMI Awards and \" Favorite Song \" at the Blockbuster Entertainment Awards , where Carey also won the award for \" Top Pop Female . \" In 1996 , Carey won many awards at the Billboard Music Awards , including \" Hot 100 Singles Artist of the Year \" , \" Hot 100 Airplay ( Always Be My Baby ) , \" \" Hot Adult Contemporary Artist of the Year \" and \" Special Award for 16 weeks at # 1 for ' One Sweet Day . ' \" \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
} |
Daydream was proven to be one of the best @-@ selling and most acclaimed albums of 1995 . When the Grammy Award nominees were announced , and Daydream was nominated for six different awards , critics began raving how it would be " cleaning up " that year . The 38th Annual Grammy Awards were held on February 28 , 1996 at the Shrine Auditorium in Los Angeles . Carey , being a multiple award nominee , was one of the headlining performers . Together with Boyz II Men , she sang a live rendition of " One Sweet Day , " to a very positive response . However , as the award winners were announced one by one , Carey watched as her name was not called up even once . Daydream had lost all of its six nominations , shocking most critics who branded it the " album of the year " . With every passing loss , the television cameras continued to zoom on Carey 's face , who was finding it more difficult to retain her smile . By the end of the night , Carey had not won a single award . The disappointment on her face was painfully obvious . While Carey was nominated again the following year , she did not perform again until the 2006 ceremony , when she was nominated for eight awards ( winning three ) for The Emancipation of Mimi .
| [
{
"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 Daydream was proven to be one of the best @-@ selling and most acclaimed albums of 1995 . When the Grammy Award nominees were announced , and Daydream was nominated for six different awards , critics began raving how it would be \" cleaning up \" that year . The 38th Annual Grammy Awards were held on February 28 , 1996 at the Shrine Auditorium in Los Angeles . Carey , being a multiple award nominee , was one of the headlining performers . Together with Boyz II Men , she sang a live rendition of \" One Sweet Day , \" to a very positive response . However , as the award winners were announced one by one , Carey watched as her name was not called up even once . Daydream had lost all of its six nominations , shocking most critics who branded it the \" album of the year \" . With every passing loss , the television cameras continued to zoom on Carey 's face , who was finding it more difficult to retain her smile . By the end of the night , Carey had not won a single award . The disappointment on her face was painfully obvious . While Carey was nominated again the following year , she did not perform again until the 2006 ceremony , when she was nominated for eight awards ( winning three ) for The Emancipation of Mimi . \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
} |
Carey directed the music video for " Fantasy " . Additionally , she chose the concept and created the idea . The video featured Carey in roller @-@ blades , blading through a theme @-@ park , while enjoying different rides and roller coasters . The video then cut into scenes of Carey dancing on top of a car , celebrating an event with many friends . Carey said her inspiration for the video was to give off a " free and open feeling , " trying to portray the freedom she had finally achieved in being allowed to direct her first video . The video for " Fantasy " debuted on September 7 , at the " MTV Video Music Awards . " Carey expressed how much she enjoyed the video 's filming , speaking about the " roller @-@ coaster scenes " :
| [
{
"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 Carey directed the music video for \" Fantasy \" . Additionally , she chose the concept and created the idea . The video featured Carey in roller @-@ blades , blading through a theme @-@ park , while enjoying different rides and roller coasters . The video then cut into scenes of Carey dancing on top of a car , celebrating an event with many friends . Carey said her inspiration for the video was to give off a \" free and open feeling , \" trying to portray the freedom she had finally achieved in being allowed to direct her first video . The video for \" Fantasy \" debuted on September 7 , at the \" MTV Video Music Awards . \" Carey expressed how much she enjoyed the video 's filming , speaking about the \" roller @-@ coaster scenes \" : \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 did not expect me to get that shot ! They were saying , ' How 's she going to sing on a roller coaster ? ... We put a little speaker on the bottom of the car , where my feet where . We built the rig in front of the roller coaster and the lens kept falling off ! "
| [
{
"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 did not expect me to get that shot ! They were saying , ' How 's she going to sing on a roller coaster ? ... We put a little speaker on the bottom of the car , where my feet where . We built the rig in front of the roller coaster and the lens kept falling off ! \" \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
} |
When Carey and Boyz II Men got together to record " One Sweet Day , " they didn 't have enough time to re @-@ unite and film a video . For this reason , a filming crew was present during the song 's recording , and filmed bits of Carey and Boyz recording the song . In an interview with Fred Bronson , Walter Afanasieff made the following statements regarding the video for " One Sweet Day " :
| [
{
"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 When Carey and Boyz II Men got together to record \" One Sweet Day , \" they didn 't have enough time to re @-@ unite and film a video . For this reason , a filming crew was present during the song 's recording , and filmed bits of Carey and Boyz recording the song . In an interview with Fred Bronson , Walter Afanasieff made the following statements regarding the video for \" One Sweet Day \" : \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
} |
" It was crazy ! They had film crews and video guys , while I 'm at the board trying to produce . And these guys were running around having a ball , because Mariah and them are laughing and screaming and they 're being interviewed . And I 'm tapping people on the shoulder . " We 've got to get to the microphone ! " They 're gone in a couple of hours , so I recorded everything they did , praying that it was enough . "
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n \" It was crazy ! They had film crews and video guys , while I 'm at the board trying to produce . And these guys were running around having a ball , because Mariah and them are laughing and screaming and they 're being interviewed . And I 'm tapping people on the shoulder . \" We 've got to get to the microphone ! \" They 're gone in a couple of hours , so I recorded everything they did , praying that it was enough . \" \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 video for " Always Be My Baby " was once again filmed by Carey . It featured Carey swinging on a swing in the middle of a dark meadow , showing her frolicking through the woodlands . The video was set as a " peaceful and relaxing " setting , in order to try to reflect the sweet and mellow song 's message . " Forever " , the album 's fifth single , was the final song to be accompanied by a music video . Carey 's label used footage from her live performances of the song in New York City and Japan and compiled them into a video . On February 11 , 2012 , Carey revealed through her Twitter account that a music video for " Underneath the Stars " was actually recorded . Filming sessions occurred in England and France ; Carey commented , however , that it " never got released and I don 't know where it is ! "
| [
{
"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 video for \" Always Be My Baby \" was once again filmed by Carey . It featured Carey swinging on a swing in the middle of a dark meadow , showing her frolicking through the woodlands . The video was set as a \" peaceful and relaxing \" setting , in order to try to reflect the sweet and mellow song 's message . \" Forever \" , the album 's fifth single , was the final song to be accompanied by a music video . Carey 's label used footage from her live performances of the song in New York City and Japan and compiled them into a video . On February 11 , 2012 , Carey revealed through her Twitter account that a music video for \" Underneath the Stars \" was actually recorded . Filming sessions occurred in England and France ; Carey commented , however , that it \" never got released and I don 't know where it is ! \" \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
} |
Leg before wicket ( lbw ) is one of the ways in which a batsman can be dismissed in the sport of cricket . Following an appeal by the fielding side , the umpire may rule a batsman out lbw if the ball would have struck the wicket , but was instead intercepted by any part of the batsman 's body ( except the hand holding the bat ) . The umpire 's decision will depend on a number of criteria , including where the ball pitched , whether the ball hit in line with the wickets , and whether the batsman was attempting to hit the ball .
| [
{
"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 Leg before wicket ( lbw ) is one of the ways in which a batsman can be dismissed in the sport of cricket . Following an appeal by the fielding side , the umpire may rule a batsman out lbw if the ball would have struck the wicket , but was instead intercepted by any part of the batsman 's body ( except the hand holding the bat ) . The umpire 's decision will depend on a number of criteria , including where the ball pitched , whether the ball hit in line with the wickets , and whether the batsman was attempting to hit the ball . \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|>Leg before wicket ( lbw ) is one of the ways in which a batsman can be dismissed in<|end-PI|> <|retrieval|> <|start-KFI|>the sport of cricket<|end-KFI|> . <|start-PI|>Following an appeal by the fielding side , the umpire may rule a batsman out lbw if the ball would have struck<|end-PI|> <|retrieval|> <|start-KFI|>the wicket , but was instead intercepted by any part of the batsman 's body ( except the hand holding the bat )<|end-KFI|> . <|start-PI|>The umpire 's decision will depend on a number of criteria , including<|end-PI|> <|retrieval|> <|start-KFI|>where the ball pitched , whether the ball hit in line with the wickets , and whether the batsman was attempting to hit the ball<|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the concept of lbw and sets up a query about which sport it belongs to, while the KFI provides the specific sport. The <|retrieval|> token is inserted to indicate where additional information about the sport might be needed.",
"Ideal Knowledge": "A comprehensive explanation of cricket dismissal methods, with a focus on leg before wicket (lbw) and its specific application in the sport of cricket.",
"Key Factual Information": "the sport of cricket",
"Prefix reformulated": "What sport is leg before wicket (lbw) used as a method of dismissal?",
"Prefix-intent": "Leg before wicket ( lbw ) is one of the ways in which a batsman can be dismissed in",
"Verification": "The KFI directly provides the sport in which lbw is used, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context for an lbw decision, and the KFI provides the specific conditions that must be met. The <|retrieval|> token is inserted to indicate where additional information about the lbw rule might be needed.",
"Ideal Knowledge": "Detailed explanation of the lbw rule in cricket, including the specific conditions that must be met for an umpire to rule a batsman out, with emphasis on the ball's trajectory and the parts of the batsman's body involved.",
"Key Factual Information": "the wicket , but was instead intercepted by any part of the batsman 's body ( except the hand holding the bat )",
"Prefix reformulated": "What condition must be met for an umpire to rule a batsman out lbw?",
"Prefix-intent": "Following an appeal by the fielding side , the umpire may rule a batsman out lbw if the ball would have struck",
"Verification": "The KFI directly provides the conditions for an lbw dismissal, as prompted by the PI."
},
{
"Explanation": "The PI introduces the idea that multiple criteria are considered in an lbw decision, and the KFI lists these specific criteria. The <|retrieval|> token is inserted to indicate where additional information about these criteria might be needed.",
"Ideal Knowledge": "A comprehensive list of criteria used by cricket umpires when making lbw decisions, including detailed explanations of each criterion and how they affect the decision-making process.",
"Key Factual Information": "where the ball pitched , whether the ball hit in line with the wickets , and whether the batsman was attempting to hit the ball",
"Prefix reformulated": "What criteria does the umpire consider when making an lbw decision?",
"Prefix-intent": "The umpire 's decision will depend on a number of criteria , including",
"Verification": "The KFI directly lists the criteria considered by the umpire, as prompted by the PI."
}
]
} |
Leg before wicket first appeared in the laws of cricket in 1774 , as batsmen began to use their pads to prevent the ball hitting their wicket . Over several years , refinements were made to clarify where the ball should pitch and to remove the element of interpreting the batsman 's intentions . The 1839 version of the law used a wording that remained in place for nearly 100 years . However , from the latter part of the 19th century , batsmen became increasingly expert at " pad @-@ play " to reduce the risk of their dismissal . Following a number of failed proposals for reform , in 1935 the law was expanded , such that batsmen could be dismissed lbw even if the ball pitched outside the line of off stump . Critics felt this change made the game unattractive as it encouraged negative tactics at the expense of leg spin bowling .
| [
{
"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 Leg before wicket first appeared in the laws of cricket in 1774 , as batsmen began to use their pads to prevent the ball hitting their wicket . Over several years , refinements were made to clarify where the ball should pitch and to remove the element of interpreting the batsman 's intentions . The 1839 version of the law used a wording that remained in place for nearly 100 years . However , from the latter part of the 19th century , batsmen became increasingly expert at \" pad @-@ play \" to reduce the risk of their dismissal . Following a number of failed proposals for reform , in 1935 the law was expanded , such that batsmen could be dismissed lbw even if the ball pitched outside the line of off stump . Critics felt this change made the game unattractive as it encouraged negative tactics at the expense of leg spin bowling . \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 considerable debate and various experiments , the law was changed again in 1972 . In an attempt to reduce pad @-@ play the new version , which is used to this day , allowed batsmen to be out lbw in some circumstances if they did not attempt to hit the ball with their bat . Since the 1990s , the availability of television replays and , later , ball @-@ tracking technology to assist umpires has increased the percentage of lbws in major matches . However , the accuracy of the technology and the consequences of its use remain controversial .
| [
{
"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 considerable debate and various experiments , the law was changed again in 1972 . In an attempt to reduce pad @-@ play the new version , which is used to this day , allowed batsmen to be out lbw in some circumstances if they did not attempt to hit the ball with their bat . Since the 1990s , the availability of television replays and , later , ball @-@ tracking technology to assist umpires has increased the percentage of lbws in major matches . However , the accuracy of the technology and the consequences of its use remain controversial . \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 his 1995 survey of cricket laws , Gerald Brodribb states : " No dismissal has produced so much argument as lbw ; it has caused trouble from its earliest days " . Owing to its complexity , the law is widely misunderstood among the general public and has proven controversial among spectators , administrators and commentators ; lbw decisions have sometimes caused crowd trouble . Since the law 's introduction , the proportion of lbw dismissals has risen steadily through the years . Statistics reveal that the probability of a batsman being dismissed lbw in a Test match varies depending on where the match is played and which teams are playing .
| [
{
"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 his 1995 survey of cricket laws , Gerald Brodribb states : \" No dismissal has produced so much argument as lbw ; it has caused trouble from its earliest days \" . Owing to its complexity , the law is widely misunderstood among the general public and has proven controversial among spectators , administrators and commentators ; lbw decisions have sometimes caused crowd trouble . Since the law 's introduction , the proportion of lbw dismissals has risen steadily through the years . Statistics reveal that the probability of a batsman being dismissed lbw in a Test match varies depending on where the match is played and which teams are playing . \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 definition of leg before wicket ( lbw ) is currently Law 36 in the Laws of Cricket , written by the Marylebone Cricket Club ( MCC ) . Before a batsman can be dismissed lbw , the fielding team must appeal to the umpire . For the batsman to be adjudged lbw , the ball , if it bounces , must pitch in line with the wickets , or on the off side of the stumps . Then the ball must strike part of the batsman 's body , without first touching his bat , in line with the wickets and have been going on to hit the stumps . The batsman may also be out lbw if , having made no attempt to hit the ball with his bat , he is struck outside the line of off stump by a ball that would have hit the wickets . The umpire must assume that the ball would have continued on the same trajectory after striking the batsman , even if it would have bounced before hitting the stumps . However , if the bowler delivers a no ball — an illegal delivery — the batsman cannot be lbw under any circumstances .
| [
{
"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 definition of leg before wicket ( lbw ) is currently Law 36 in the Laws of Cricket , written by the Marylebone Cricket Club ( MCC ) . Before a batsman can be dismissed lbw , the fielding team must appeal to the umpire . For the batsman to be adjudged lbw , the ball , if it bounces , must pitch in line with the wickets , or on the off side of the stumps . Then the ball must strike part of the batsman 's body , without first touching his bat , in line with the wickets and have been going on to hit the stumps . The batsman may also be out lbw if , having made no attempt to hit the ball with his bat , he is struck outside the line of off stump by a ball that would have hit the wickets . The umpire must assume that the ball would have continued on the same trajectory after striking the batsman , even if it would have bounced before hitting the stumps . However , if the bowler delivers a no ball — an illegal delivery — the batsman cannot be lbw under any circumstances . \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 batsman can be out lbw even if the ball did not hit his leg : for example , a batsman struck on the head could be lbw . However , he cannot be lbw if the ball pitches on the leg side of the stumps ( " outside leg stump " ) , even if the ball would have otherwise hit the wickets . Similarly , a batsman who has attempted to hit the ball with his bat cannot be lbw if the ball strikes him outside the line of off stump . However , some shots in cricket , such as the switch hit or reverse sweep , involve the batsman switching between a right- and left @-@ handed stance ; this affects the location of the off and leg side , which are determined by the stance . The law explicitly states that the off side is determined by the batsman 's position when the bowler commences his run @-@ 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 A batsman can be out lbw even if the ball did not hit his leg : for example , a batsman struck on the head could be lbw . However , he cannot be lbw if the ball pitches on the leg side of the stumps ( \" outside leg stump \" ) , even if the ball would have otherwise hit the wickets . Similarly , a batsman who has attempted to hit the ball with his bat cannot be lbw if the ball strikes him outside the line of off stump . However , some shots in cricket , such as the switch hit or reverse sweep , involve the batsman switching between a right- and left @-@ handed stance ; this affects the location of the off and leg side , which are determined by the stance . The law explicitly states that the off side is determined by the batsman 's position when the bowler commences his run @-@ 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": "<|start-PI|>A batsman can be out lbw even if the ball did not hit his leg : for example<|end-PI|> <|retrieval|> <|start-KFI|>a batsman struck on the head could be lbw<|end-KFI|>. <|start-PI|>However , he cannot be lbw if the ball pitches on the leg side of the stumps<|end-PI|> <|retrieval|> <|start-KFI|>( \" outside leg stump \" ) , even if the ball would have otherwise hit the wickets<|end-KFI|>. <|start-PI|>Similarly , a batsman who has attempted to hit the ball with his bat cannot be lbw if the ball strikes him<|end-PI|> <|retrieval|> <|start-KFI|>outside the line of off stump<|end-KFI|>. However , some shots in cricket , such as the switch hit or reverse sweep , involve the batsman switching between a right- and left @-@ handed stance ; this affects the location of the off and leg side , which are determined by the stance . <|start-PI|>The law explicitly states that the off side is determined by<|end-PI|> <|retrieval|> <|start-KFI|>the batsman 's position when the bowler commences his run @-@ up<|end-KFI|>.",
"none": null,
"results": [
{
"Explanation": "The PI introduces the concept of lbw and sets up a query about unusual scenarios, while the KFI provides a specific example. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.",
"Ideal Knowledge": "A comprehensive explanation of the lbw (leg before wicket) rule in cricket, including various scenarios where a batsman can be given out lbw, even when the ball doesn't hit the leg.",
"Key Factual Information": "a batsman struck on the head could be lbw",
"Prefix reformulated": "In what scenario can a batsman be out lbw even if the ball did not hit his leg?",
"Prefix-intent": "A batsman can be out lbw even if the ball did not hit his leg : for example",
"Verification": "The KFI directly provides an example of when a batsman can be out lbw without the ball hitting his leg, as prompted by the PI."
},
{
"Explanation": "The PI introduces a limitation to the lbw rule, and the KFI provides the specific term and additional context. The <|retrieval|> token is inserted to indicate where additional information about cricket rules might be needed.",
"Ideal Knowledge": "Detailed explanation of the lbw rule in cricket, focusing on the 'outside leg stump' exception and its implications for dismissals.",
"Key Factual Information": "( \" outside leg stump \" ) , even if the ball would have otherwise hit the wickets",
"Prefix reformulated": "When can't a batsman be given out lbw in cricket?",
"Prefix-intent": "However , he cannot be lbw if the ball pitches on the leg side of the stumps",
"Verification": "The KFI directly provides the specific condition when a batsman cannot be given out lbw, as prompted by the PI."
},
{
"Explanation": "The PI sets up another limitation to the lbw rule, and the KFI provides the specific location. The <|retrieval|> token is inserted to indicate where additional information about cricket rules might be needed.",
"Ideal Knowledge": "Detailed explanation of the lbw rule in cricket, focusing on the 'outside the line of off stump' exception when the batsman has attempted to play a shot, and its implications for dismissals.",
"Key Factual Information": "outside the line of off stump",
"Prefix reformulated": "Where can't a batsman be given out lbw if he has attempted to hit the ball?",
"Prefix-intent": "Similarly , a batsman who has attempted to hit the ball with his bat cannot be lbw if the ball strikes him",
"Verification": "The KFI directly provides the specific location where a batsman cannot be given out lbw if he has attempted to hit the ball, as prompted by the PI."
},
{
"Explanation": "The PI introduces a specific rule about determining the off side, and the KFI provides the exact condition. The <|retrieval|> token is inserted to indicate where additional information about cricket rules might be needed.",
"Ideal Knowledge": "Detailed explanation of how the off side and leg side are determined in cricket, including the importance of the batsman's stance at the moment the bowler begins their run-up.",
"Key Factual Information": "the batsman 's position when the bowler commences his run @-@ up",
"Prefix reformulated": "How is the off side determined in cricket according to the law?",
"Prefix-intent": "The law explicitly states that the off side is determined by",
"Verification": "The KFI directly provides the specific condition that determines the off side in cricket, as prompted by the PI."
}
]
} |
According to MCC guidelines for umpires , factors to consider when giving an lbw decision include the angle at which the ball was travelling and whether the ball was swinging through the air . He must also account for the height of the ball at impact and how far from the wicket the batsman was standing ; from this information he must determine if the ball would have passed over the stumps or struck them . The MCC guidance states that it is easier to make a decision when the ball strikes the batsman without pitching , but that the difficulty increases when the ball has bounced and more so when there is a shorter time between the ball pitching and striking the batsman .
| [
{
"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 According to MCC guidelines for umpires , factors to consider when giving an lbw decision include the angle at which the ball was travelling and whether the ball was swinging through the air . He must also account for the height of the ball at impact and how far from the wicket the batsman was standing ; from this information he must determine if the ball would have passed over the stumps or struck them . The MCC guidance states that it is easier to make a decision when the ball strikes the batsman without pitching , but that the difficulty increases when the ball has bounced and more so when there is a shorter time between the ball pitching and striking the batsman . \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 earliest known written version of the Laws of Cricket , dating from 1744 , does not include an lbw rule . At the time , batsmen in English cricket used curved bats , which made it unlikely that they would be able to stand directly in front of the wickets . However , a clause in the 1744 laws gave umpires the power to take action if the batsman was " standing unfair to strike " . Cricket bats were modified to become straighter over the following years , allowing batsmen to stand closer to the wickets . Subsequently , some players deliberately began to obstruct the ball from hitting the wickets . Such tactics were criticised by writers and a revision of the laws in 1774 ruled that the batsman was out if he deliberately stopped the ball from hitting the wicket with his leg . However , critics noted that the umpires were left the difficult task of interpreting the intentions of batsmen . The 1788 version of the laws no longer required the umpires to take account of the batsman 's intent ; now a batsman was lbw if he stopped a ball that " pitch [ ed ] straight " . Further clarification of the law came in 1823 , when a condition was added that " the ball must be delivered in a straight line to the wicket " . The ambiguity of the wording was highlighted when two prominent umpires disagreed over whether the ball had to travel in a straight line from the bowler to the wicket , or between the wickets at either end of the pitch . In 1839 the MCC , by then responsible for drafting the Laws of Cricket , endorsed the latter interpretation and ruled the batsman out lbw if the ball pitched in between the wickets and would have hit the stumps .
| [
{
"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 earliest known written version of the Laws of Cricket , dating from 1744 , does not include an lbw rule . At the time , batsmen in English cricket used curved bats , which made it unlikely that they would be able to stand directly in front of the wickets . However , a clause in the 1744 laws gave umpires the power to take action if the batsman was \" standing unfair to strike \" . Cricket bats were modified to become straighter over the following years , allowing batsmen to stand closer to the wickets . Subsequently , some players deliberately began to obstruct the ball from hitting the wickets . Such tactics were criticised by writers and a revision of the laws in 1774 ruled that the batsman was out if he deliberately stopped the ball from hitting the wicket with his leg . However , critics noted that the umpires were left the difficult task of interpreting the intentions of batsmen . The 1788 version of the laws no longer required the umpires to take account of the batsman 's intent ; now a batsman was lbw if he stopped a ball that \" pitch [ ed ] straight \" . Further clarification of the law came in 1823 , when a condition was added that \" the ball must be delivered in a straight line to the wicket \" . The ambiguity of the wording was highlighted when two prominent umpires disagreed over whether the ball had to travel in a straight line from the bowler to the wicket , or between the wickets at either end of the pitch . In 1839 the MCC , by then responsible for drafting the Laws of Cricket , endorsed the latter interpretation and ruled the batsman out lbw if the ball pitched in between the wickets and would have hit the stumps . \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 essence , the lbw law remained the same between 1839 and 1937 , despite several campaigns to have it changed . An 1863 proposal to allow a batsman to be lbw if the ball hit his body at any point between the wickets , regardless of where the ball pitched or whether it would hit the wicket at all , came to nothing . There were few complaints until the proportion of lbw dismissals in county cricket began to increase during the 1880s . Until then , batsmen used their pads only to protect their legs ; their use for any other purposes was considered unsporting , and some amateur cricketers did not wear them at all . As cricket became more organised and competitive , some batsmen began to use their pads as a second line of defence : they lined them up with the ball so that if they missed with the bat , the ball struck the pad instead of the wicket . Some players took this further ; if the delivery was not an easy one from which to score runs , they attempted no shot and allowed the ball to bounce safely off their pads . Arthur Shrewsbury was the first prominent player to use such methods , and others followed . Criticism of this practice was heightened by the increased quality and reliability of cricket pitches , which made batting easier , led to higher scores and created a perceived imbalance in the game .
| [
{
"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 essence , the lbw law remained the same between 1839 and 1937 , despite several campaigns to have it changed . An 1863 proposal to allow a batsman to be lbw if the ball hit his body at any point between the wickets , regardless of where the ball pitched or whether it would hit the wicket at all , came to nothing . There were few complaints until the proportion of lbw dismissals in county cricket began to increase during the 1880s . Until then , batsmen used their pads only to protect their legs ; their use for any other purposes was considered unsporting , and some amateur cricketers did not wear them at all . As cricket became more organised and competitive , some batsmen began to use their pads as a second line of defence : they lined them up with the ball so that if they missed with the bat , the ball struck the pad instead of the wicket . Some players took this further ; if the delivery was not an easy one from which to score runs , they attempted no shot and allowed the ball to bounce safely off their pads . Arthur Shrewsbury was the first prominent player to use such methods , and others followed . Criticism of this practice was heightened by the increased quality and reliability of cricket pitches , which made batting easier , led to higher scores and created a perceived imbalance in the game . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "In essence, <|start-PI|>the lbw law remained the same between<|end-PI|> <|retrieval|> <|start-KFI|>1839 and 1937, despite several campaigns to have it changed<|end-KFI|>. <|start-PI|>An 1863 proposal to allow a batsman to be lbw if<|end-PI|> <|retrieval|> <|start-KFI|>the ball hit his body at any point between the wickets, regardless of where the ball pitched or whether it would hit the wicket at all, came to nothing<|end-KFI|>. There were few complaints until the proportion of lbw dismissals in county cricket began to increase during the 1880s. <|start-PI|>Until then, batsmen used their pads only to<|end-PI|> <|retrieval|> <|start-KFI|>protect their legs; their use for any other purposes was considered unsporting, and some amateur cricketers did not wear them at all<|end-KFI|>. <|start-PI|>As cricket became more organised and competitive, some batsmen began to use their pads as<|end-PI|> <|retrieval|> <|start-KFI|>a second line of defence: they lined them up with the ball so that if they missed with the bat, the ball struck the pad instead of the wicket<|end-KFI|>. Some players took this further; if the delivery was not an easy one from which to score runs, they attempted no shot and allowed the ball to bounce safely off their pads. <|start-PI|>Arthur Shrewsbury was the first prominent player to<|end-PI|> <|retrieval|> <|start-KFI|>use such methods, and others followed<|end-KFI|>. Criticism of this practice was heightened by the increased quality and reliability of cricket pitches, which made batting easier, led to higher scores and created a perceived imbalance in the game.",
"none": null,
"results": [
{
"Explanation": "The PI introduces the topic of the lbw law's stability, while the KFI provides the specific time frame and additional context. The <|retrieval|> token is inserted to indicate where additional historical information might be needed.",
"Ideal Knowledge": "A detailed timeline of cricket law changes, specifically focusing on the leg before wicket (lbw) rule and its stability between 1839 and 1937, including information about unsuccessful campaigns for change during this period.",
"Key Factual Information": "1839 and 1937, despite several campaigns to have it changed",
"Prefix reformulated": "What was the time period during which the lbw law remained unchanged?",
"Prefix-intent": "In essence, the lbw law remained the same between",
"Verification": "The KFI directly provides the time period requested in the PI."
},
{
"Explanation": "The PI introduces a specific proposal from 1863, and the KFI provides the details of the proposed change and its outcome. The <|retrieval|> token is inserted to indicate where additional historical context might be needed.",
"Ideal Knowledge": "Detailed information about the 1863 proposal to change the lbw rule in cricket, including the specific suggested modifications and the reasons why the proposal was not adopted.",
"Key Factual Information": "the ball hit his body at any point between the wickets, regardless of where the ball pitched or whether it would hit the wicket at all, came to nothing",
"Prefix reformulated": "What was proposed in 1863 regarding the lbw rule in cricket?",
"Prefix-intent": "An 1863 proposal to allow a batsman to be lbw if",
"Verification": "The KFI directly describes the proposed change to the lbw rule, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the original use of pads in cricket, and the KFI provides the specific purpose and additional cultural context. The <|retrieval|> token is inserted to indicate where additional historical information might be needed.",
"Ideal Knowledge": "Information on the historical use of protective equipment in cricket, specifically focusing on the introduction and evolution of leg pads, including their initial purpose and the attitudes towards their use in early cricket.",
"Key Factual Information": "protect their legs; their use for any other purposes was considered unsporting, and some amateur cricketers did not wear them at all",
"Prefix reformulated": "What was the original purpose of pads for batsmen in cricket?",
"Prefix-intent": "Until then, batsmen used their pads only to",
"Verification": "The KFI directly answers the question about the original purpose of pads and provides additional context, as prompted by the PI."
},
{
"Explanation": "The PI introduces the changing nature of cricket and sets up a query about pad usage, while the KFI provides the specific new strategy employed by batsmen. The <|retrieval|> token is inserted to indicate where additional information about cricket tactics might be needed.",
"Ideal Knowledge": "Detailed information on the evolution of batting techniques in cricket, specifically focusing on the strategic use of pads as a defensive tool and how this practice developed as the sport became more competitive.",
"Key Factual Information": "a second line of defence: they lined them up with the ball so that if they missed with the bat, the ball struck the pad instead of the wicket",
"Prefix reformulated": "How did batsmen's use of pads change as cricket became more organized and competitive?",
"Prefix-intent": "As cricket became more organised and competitive, some batsmen began to use their pads as",
"Verification": "The KFI directly describes how the use of pads changed, as prompted by the PI."
},
{
"Explanation": "The PI introduces Arthur Shrewsbury and sets up a query about his role, while the KFI confirms his pioneering status in using the new pad techniques. The <|retrieval|> token is inserted to indicate where additional biographical information might be needed.",
"Ideal Knowledge": "Biographical information on Arthur Shrewsbury, focusing on his cricket career and his role in popularizing the strategic use of pads as a defensive technique in batting.",
"Key Factual Information": "use such methods, and others followed",
"Prefix reformulated": "What innovative technique did Arthur Shrewsbury introduce in cricket?",
"Prefix-intent": "Arthur Shrewsbury was the first prominent player to",
"Verification": "The KFI directly states that Shrewsbury was the first to use the previously described methods, as prompted by the PI."
}
]
} |
Several proposals were made to prevent pad @-@ play . At a meeting of representatives of the main county cricket clubs in 1888 , one representative expressed the opinion that a " batsman who defended his wicket with his body instead of with his bat should be punished " . The representatives supported a motion to alter the law to state that the batsman would be out if he stopped a ball that would have hit the wicket ; in contrast to the existing wording , this took no account of where the ball pitched relative to the wickets . Further proposals included one in which the intent of the batsman was taken into account , but no laws were changed and the MCC merely issued a condemnation of the practice of using pads for defence . This reduced pad @-@ play for a short time , but when it increased again , a second pronouncement by the MCC had little effect .
| [
{
"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 proposals were made to prevent pad @-@ play . At a meeting of representatives of the main county cricket clubs in 1888 , one representative expressed the opinion that a \" batsman who defended his wicket with his body instead of with his bat should be punished \" . The representatives supported a motion to alter the law to state that the batsman would be out if he stopped a ball that would have hit the wicket ; in contrast to the existing wording , this took no account of where the ball pitched relative to the wickets . Further proposals included one in which the intent of the batsman was taken into account , but no laws were changed and the MCC merely issued a condemnation of the practice of using pads for defence . This reduced pad @-@ play for a short time , but when it increased again , a second pronouncement by the MCC had little effect . \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
} |
Further discussion on altering the law took place in 1899 , when several prominent cricketers supported an amendment similar to the 1888 proposal : the batsman would be out if the ball would have hit the wicket , where it pitched was irrelevant . At a Special General Meeting of the MCC in 1902 , Alfred Lyttelton formally proposed this amendment ; the motion was supported by 259 votes to 188 , but failed to secure the two @-@ thirds majority required to change the laws . A. G. Steel was the principal opponent of the change , as he believed it would make the task of the umpires too difficult , but he later regretted his stance . Lyttelton 's brother , Robert , supported the alteration and campaigned for the rest of his life to have the lbw law altered . As evidence that pad @-@ play was increasing and needed to be curtailed , he cited the growing number of wickets which were falling lbw : the proportion rose from 2 % of dismissals in 1870 to 6 % in 1890 , and 12 % in 1923 . In 1902 , the proposed new law was tried in the Minor Counties Championship , but deemed a failure . An increase in the size of the stumps was one of several other rejected proposals at this time to reduce the dominance of batsmen over bowlers .
| [
{
"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 Further discussion on altering the law took place in 1899 , when several prominent cricketers supported an amendment similar to the 1888 proposal : the batsman would be out if the ball would have hit the wicket , where it pitched was irrelevant . At a Special General Meeting of the MCC in 1902 , Alfred Lyttelton formally proposed this amendment ; the motion was supported by 259 votes to 188 , but failed to secure the two @-@ thirds majority required to change the laws . A. G. Steel was the principal opponent of the change , as he believed it would make the task of the umpires too difficult , but he later regretted his stance . Lyttelton 's brother , Robert , supported the alteration and campaigned for the rest of his life to have the lbw law altered . As evidence that pad @-@ play was increasing and needed to be curtailed , he cited the growing number of wickets which were falling lbw : the proportion rose from 2 % of dismissals in 1870 to 6 % in 1890 , and 12 % in 1923 . In 1902 , the proposed new law was tried in the Minor Counties Championship , but deemed a failure . An increase in the size of the stumps was one of several other rejected proposals at this time to reduce the dominance of batsmen over bowlers . \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": "Further discussion on altering the law took place in 1899 , when <|start-PI|> several prominent cricketers supported an amendment similar to the 1888 proposal : <|end-PI|> <|retrieval|> <|start-KFI|> the batsman would be out if the ball would have hit the wicket , where it pitched was irrelevant <|end-KFI|> . <|start-PI|> At a Special General Meeting of the MCC in 1902 , Alfred Lyttelton formally proposed this amendment ; the motion was supported by <|end-PI|> <|retrieval|> <|start-KFI|> 259 votes to 188 , but failed to secure the two @-@ thirds majority required to change the laws <|end-KFI|> . A. G. Steel was the principal opponent of the change , as he believed it would make the task of the umpires too difficult , but he later regretted his stance . Lyttelton 's brother , Robert , supported the alteration and campaigned for the rest of his life to have the lbw law altered . <|start-PI|> As evidence that pad @-@ play was increasing and needed to be curtailed , he cited the growing number of wickets which were falling lbw : the proportion rose from <|end-PI|> <|retrieval|> <|start-KFI|> 2 % of dismissals in 1870 to 6 % in 1890 , and 12 % in 1923 <|end-KFI|> . In 1902 , the proposed new law was tried in the Minor Counties Championship , but deemed a failure . An increase in the size of the stumps was one of several other rejected proposals at this time to reduce the dominance of batsmen over bowlers .",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of a discussion about changing cricket laws, and the KFI provides the specific details of the proposed amendment. The <|retrieval|> token is inserted to indicate where additional historical information might be needed.",
"Ideal Knowledge": "A detailed history of cricket law amendments, specifically focusing on the 1888 proposal and the 1899 discussion, including the proposed changes to the lbw (leg before wicket) rule.",
"Key Factual Information": "the batsman would be out if the ball would have hit the wicket , where it pitched was irrelevant",
"Prefix reformulated": "What amendment did prominent cricketers support in 1899 regarding cricket law?",
"Prefix-intent": "Further discussion on altering the law took place in 1899 , when several prominent cricketers supported an amendment similar to the 1888 proposal :",
"Verification": "The KFI directly provides the details of the amendment supported by cricketers in 1899, as prompted by the PI."
},
{
"Explanation": "The PI introduces the formal proposal at a specific meeting, and the KFI provides the voting results and outcome. The <|retrieval|> token is inserted to indicate where additional information about the meeting and voting process might be needed.",
"Ideal Knowledge": "Detailed records of the MCC Special General Meeting in 1902, including the full voting results, the specific requirements for changing laws, and the context of Alfred Lyttelton's proposal.",
"Key Factual Information": "259 votes to 188 , but failed to secure the two @-@ thirds majority required to change the laws",
"Prefix reformulated": "What was the voting result for Alfred Lyttelton's proposed amendment at the MCC Special General Meeting in 1902?",
"Prefix-intent": "At a Special General Meeting of the MCC in 1902 , Alfred Lyttelton formally proposed this amendment ; the motion was supported by",
"Verification": "The KFI directly provides the voting results and outcome of the proposed amendment, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of increasing pad-play and its impact on lbw dismissals, and the KFI provides specific statistical evidence. The <|retrieval|> token is inserted to indicate where additional statistical information might be needed.",
"Ideal Knowledge": "Comprehensive cricket statistics from 1870 to 1923, focusing on the types of dismissals, particularly the increase in lbw (leg before wicket) dismissals, with year-by-year data and analysis of trends.",
"Key Factual Information": "2 % of dismissals in 1870 to 6 % in 1890 , and 12 % in 1923",
"Prefix reformulated": "How did the proportion of lbw dismissals change from 1870 to 1923?",
"Prefix-intent": "As evidence that pad @-@ play was increasing and needed to be curtailed , he cited the growing number of wickets which were falling lbw : the proportion rose from",
"Verification": "The KFI directly provides the statistical changes in lbw dismissals over time, as prompted by the PI."
}
]
} |
Between 1900 and the 1930s , the number of runs scored by batsmen , and the proportion of lbw dismissals , continued to rise . Bowlers grew increasingly frustrated with pad @-@ play and the extent to which batsmen refused to play shots at bowling directed outside the off stump , simply allowing it to pass by . The English fast bowler Harold Larwood responded by targeting leg stump , frequently hitting the batsman with the ball in the process . This developed into the controversial Bodyline tactics he used in Australia in 1932 – 33 . Some batsmen began to go further and preferred to kick away balls pitched outside off stump — reaching out to kick the ball instead of allowing it to hit their pads — if they presented any threat , knowing that they could not be dismissed lbw . The authorities believed these developments represented poor entertainment value . At the height of the Bodyline controversy in 1933 , Donald Bradman , the leading Australian batsman and primary target of the English bowlers , wrote to the MCC recommending an alteration of the lbw law to create more exciting games .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Between 1900 and the 1930s , the number of runs scored by batsmen , and the proportion of lbw dismissals , continued to rise . Bowlers grew increasingly frustrated with pad @-@ play and the extent to which batsmen refused to play shots at bowling directed outside the off stump , simply allowing it to pass by . The English fast bowler Harold Larwood responded by targeting leg stump , frequently hitting the batsman with the ball in the process . This developed into the controversial Bodyline tactics he used in Australia in 1932 – 33 . Some batsmen began to go further and preferred to kick away balls pitched outside off stump — reaching out to kick the ball instead of allowing it to hit their pads — if they presented any threat , knowing that they could not be dismissed lbw . The authorities believed these developments represented poor entertainment value . At the height of the Bodyline controversy in 1933 , Donald Bradman , the leading Australian batsman and primary target of the English bowlers , wrote to the MCC recommending an alteration of the lbw law to create more exciting games . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": null,
"none": "",
"results": null
} |
To address the problem , and redress the balance for bowlers , the MCC made some alterations to the laws . The size of the ball was reduced in 1927 , and that of the stumps increased in 1931 , but the changes had little effect . Between 1929 and 1933 , county authorities conducted a trial in which a batsman could be lbw if he had hit the ball onto his pads . Then , in 1935 , an experimental law was introduced in which the batsman could be dismissed lbw even if the ball pitched outside the line of off stump — in other words , a ball that turned or swung into the batsman but did not pitch in line with the wickets . However , the ball was still required to strike the batsman in line with the wickets . The umpire signalled to the scorers when he declared a batsman out under the new rule , and any such dismissal was designated " lbw ( n ) " on the scorecard .
| [
{
"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 To address the problem , and redress the balance for bowlers , the MCC made some alterations to the laws . The size of the ball was reduced in 1927 , and that of the stumps increased in 1931 , but the changes had little effect . Between 1929 and 1933 , county authorities conducted a trial in which a batsman could be lbw if he had hit the ball onto his pads . Then , in 1935 , an experimental law was introduced in which the batsman could be dismissed lbw even if the ball pitched outside the line of off stump — in other words , a ball that turned or swung into the batsman but did not pitch in line with the wickets . However , the ball was still required to strike the batsman in line with the wickets . The umpire signalled to the scorers when he declared a batsman out under the new rule , and any such dismissal was designated \" lbw ( n ) \" on the scorecard . \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": "To address the problem, and redress the balance for bowlers, the MCC made some alterations to the laws. <|start-PI|>The size of the ball was reduced in<|end-PI|> <|retrieval|> <|start-KFI|>1927, and that of the stumps increased in 1931<|end-KFI|>, but the changes had little effect. <|start-PI|>Between 1929 and 1933, county authorities conducted a trial in which<|end-PI|> <|retrieval|> <|start-KFI|>a batsman could be lbw if he had hit the ball onto his pads<|end-KFI|>. <|start-PI|>Then, in 1935, an experimental law was introduced in which the batsman could be dismissed lbw even if<|end-PI|> <|retrieval|> <|start-KFI|>the ball pitched outside the line of off stump — in other words, a ball that turned or swung into the batsman but did not pitch in line with the wickets<|end-KFI|>. However, the ball was still required to strike the batsman in line with the wickets. The umpire signalled to the scorers when he declared a batsman out under the new rule, and any such dismissal was designated 'lbw (n)' on the scorecard.",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of rule changes, and the KFI provides specific dates for these changes. The <|retrieval|> token is inserted to indicate where additional historical information might be needed.",
"Ideal Knowledge": "A timeline of cricket rule changes implemented by the MCC, specifically focusing on alterations to ball size and stump dimensions in the late 1920s and early 1930s.",
"Key Factual Information": "1927, and that of the stumps increased in 1931",
"Prefix reformulated": "When did the MCC reduce the size of the cricket ball to address the balance between batsmen and bowlers?",
"Prefix-intent": "To address the problem, and redress the balance for bowlers, the MCC made some alterations to the laws. The size of the ball was reduced in",
"Verification": "The KFI directly provides the dates of the changes mentioned in the PI."
},
{
"Explanation": "The PI introduces a specific time period and mentions a trial, while the KFI provides details about the trial. The <|retrieval|> token is inserted to indicate where additional information about this experimental rule might be needed.",
"Ideal Knowledge": "Detailed information about experimental cricket rules implemented in county matches during the early 1930s, particularly focusing on changes to the leg before wicket (lbw) rule.",
"Key Factual Information": "a batsman could be lbw if he had hit the ball onto his pads",
"Prefix reformulated": "What trial did county cricket authorities conduct between 1929 and 1933?",
"Prefix-intent": "Between 1929 and 1933, county authorities conducted a trial in which",
"Verification": "The KFI directly describes the trial mentioned in the PI."
},
{
"Explanation": "The PI introduces a specific year and mentions an experimental law, while the KFI provides details about the law. The <|retrieval|> token is inserted to indicate where additional information about this rule change might be needed.",
"Ideal Knowledge": "A comprehensive explanation of the 1935 experimental lbw rule in cricket, including diagrams or animations showing how the new rule differed from the previous one, particularly regarding ball placement relative to the stumps.",
"Key Factual Information": "the ball pitched outside the line of off stump — in other words, a ball that turned or swung into the batsman but did not pitch in line with the wickets",
"Prefix reformulated": "What change was made to the lbw rule in 1935?",
"Prefix-intent": "Then, in 1935, an experimental law was introduced in which the batsman could be dismissed lbw even if",
"Verification": "The KFI directly explains the change in the lbw rule mentioned in the PI."
}
]
} |
Several leading batsmen opposed the new law , including the professional Herbert Sutcliffe , known as an exponent of pad @-@ play , and amateurs Errol Holmes and Bob Wyatt . Wisden Cricketers ' Almanack noted that these three improved their batting records during the 1935 season , but batsmen generally were less successful . There were also fewer drawn matches . There was an increase in the number of lbws — out of 1 @,@ 560 lbw dismissals in first @-@ class matches in 1935 , 483 were given under the amended law . Wisden judged the experiment a success and several of its opponents changed their mind by the end of the season ; batsmen soon became accustomed to the alteration . Although Australian authorities were less convinced , and did not immediately introduce the revision into domestic first @-@ class cricket , in 1937 the new rule became part of the Laws of Cricket .
| [
{
"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 leading batsmen opposed the new law , including the professional Herbert Sutcliffe , known as an exponent of pad @-@ play , and amateurs Errol Holmes and Bob Wyatt . Wisden Cricketers ' Almanack noted that these three improved their batting records during the 1935 season , but batsmen generally were less successful . There were also fewer drawn matches . There was an increase in the number of lbws — out of 1 @,@ 560 lbw dismissals in first @-@ class matches in 1935 , 483 were given under the amended law . Wisden judged the experiment a success and several of its opponents changed their mind by the end of the season ; batsmen soon became accustomed to the alteration . Although Australian authorities were less convinced , and did not immediately introduce the revision into domestic first @-@ class cricket , in 1937 the new rule became part of the Laws of Cricket . \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": "Several leading batsmen opposed the new law , including <|start-PI|> Several leading batsmen opposed the new law , including <|end-PI|> <|retrieval|> <|start-KFI|> the professional Herbert Sutcliffe , known as an exponent of pad @-@ play , and amateurs Errol Holmes and Bob Wyatt <|end-KFI|> . <|start-PI|> Wisden Cricketers ' Almanack noted that these three improved their batting records during <|end-PI|> <|retrieval|> <|start-KFI|> the 1935 season , but batsmen generally were less successful <|end-KFI|> . There were also fewer drawn matches . <|start-PI|> There was an increase in the number of lbws — out of 1 @,@ 560 lbw dismissals in first @-@ class matches in 1935 , <|end-PI|> <|retrieval|> <|start-KFI|> 483 were given under the amended law <|end-KFI|> . Wisden judged the experiment a success and several of its opponents changed their mind by the end of the season ; batsmen soon became accustomed to the alteration . <|start-PI|> Although Australian authorities were less convinced , and did not immediately introduce the revision into domestic first @-@ class cricket , in <|end-PI|> <|retrieval|> <|start-KFI|> 1937 the new rule became part of the Laws of Cricket <|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI introduces the opposition to a new law, while the KFI provides specific names and details. The <|retrieval|> token is inserted to indicate where additional information about these players might be needed.",
"Ideal Knowledge": "A list of prominent cricketers from the 1930s, their playing styles, and their stances on rule changes, particularly focusing on Herbert Sutcliffe, Errol Holmes, and Bob Wyatt.",
"Key Factual Information": "the professional Herbert Sutcliffe , known as an exponent of pad @-@ play , and amateurs Errol Holmes and Bob Wyatt",
"Prefix reformulated": "Who were the leading batsmen that opposed the new cricket law?",
"Prefix-intent": "Several leading batsmen opposed the new law , including",
"Verification": "The KFI directly lists the batsmen who opposed the new law, as prompted by the PI."
},
{
"Explanation": "The PI sets up a query about the performance of specific players, and the KFI provides the season and overall context. The <|retrieval|> token is inserted to indicate where additional statistical information might be needed.",
"Ideal Knowledge": "Statistical data from the 1935 cricket season, comparing the performance of Herbert Sutcliffe, Errol Holmes, and Bob Wyatt to their previous seasons and to other batsmen in general.",
"Key Factual Information": "the 1935 season , but batsmen generally were less successful",
"Prefix reformulated": "When did Herbert Sutcliffe, Errol Holmes, and Bob Wyatt improve their batting records according to Wisden Cricketers' Almanack?",
"Prefix-intent": "Wisden Cricketers ' Almanack noted that these three improved their batting records during",
"Verification": "The KFI directly provides the season when the improvement occurred and adds context about other batsmen, as prompted by the PI."
},
{
"Explanation": "The PI introduces the increase in lbws and sets up a query about the specific number, while the KFI provides the exact figure. The <|retrieval|> token is inserted to indicate where additional statistical information might be needed.",
"Ideal Knowledge": "Detailed statistics on lbw dismissals in first-class cricket matches during the 1935 season, comparing the numbers before and after the law amendment.",
"Key Factual Information": "483 were given under the amended law",
"Prefix reformulated": "How many lbw dismissals were given under the amended law in first-class matches in 1935?",
"Prefix-intent": "There was an increase in the number of lbws — out of 1 @,@ 560 lbw dismissals in first @-@ class matches in 1935 ,",
"Verification": "The KFI directly provides the number of lbw dismissals under the new law, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of Australian reluctance and leads to a query about when the rule was officially adopted, while the KFI provides the specific year. The <|retrieval|> token is inserted to indicate where additional historical information might be needed.",
"Ideal Knowledge": "A timeline of changes to the Laws of Cricket, focusing on the lbw rule amendments in the 1930s and the process of their adoption in different cricket-playing nations.",
"Key Factual Information": "1937 the new rule became part of the Laws of Cricket",
"Prefix reformulated": "When did the new lbw rule become part of the Laws of Cricket?",
"Prefix-intent": "Although Australian authorities were less convinced , and did not immediately introduce the revision into domestic first @-@ class cricket , in",
"Verification": "The KFI directly provides the year when the rule was officially adopted, as prompted by the PI."
}
]
} |
According to Gerald Brodribb , in his survey and history of the Laws , the change produced more " enterprising " , exciting cricket but any alteration in outlook was halted by the Second World War . When the sport resumed in 1946 , batsmen were out of practice and the amended lbw law played into the hands of off spin and inswing bowlers , who began to dominate county cricket . The cricket historian Derek Birley notes that many of these bowlers imitated the methods of Alec Bedser , an inswing bowler who was successful immediately after the war , but that the resulting cricket was unexciting to watch . The revised lbw law , and other alterations in the game in favour of the bowler , further encouraged such bowling . The new law continued to provoke debate among writers and cricketers ; many former players claimed that the alteration had caused a deterioration in batting and reduced the number of shots played on the off side . A 1963 report in The Times blamed the law for reducing the variety of bowling styles : " the change has led to a steady increase in the amount of seam and off @-@ spin bowling . Whereas in the early thirties every county had a leg spinner and an orthodox left arm spinner , leg spinners , at any rate , are now few and far between . Walk on to any of the first @-@ class grounds at any time tomorrow and the chances are that you will see the wicketkeeper standing back and a medium pace bowler in action ... there is little doubt that the game , as a spectacle , is less attractive than it was . " Several critics , including Bob Wyatt , maintained that the lbw law should be returned to its pre @-@ 1935 wording ; he campaigned to do so until his death in 1995 . On the other hand , Bradman , in the 1950s , proposed extending the law so that batsmen could be lbw even if they were struck outside the line of off stump . An MCC study of the state of cricket , carried out in 1956 and 1957 , examined the prevalent and unpopular tactic involving off @-@ spin and inswing bowlers aiming at leg stump with fielders concentrated on the leg side . Rather than alter the lbw law to combat the problem , the MCC reduced the number of fielders allowed on the leg side .
| [
{
"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 According to Gerald Brodribb , in his survey and history of the Laws , the change produced more \" enterprising \" , exciting cricket but any alteration in outlook was halted by the Second World War . When the sport resumed in 1946 , batsmen were out of practice and the amended lbw law played into the hands of off spin and inswing bowlers , who began to dominate county cricket . The cricket historian Derek Birley notes that many of these bowlers imitated the methods of Alec Bedser , an inswing bowler who was successful immediately after the war , but that the resulting cricket was unexciting to watch . The revised lbw law , and other alterations in the game in favour of the bowler , further encouraged such bowling . The new law continued to provoke debate among writers and cricketers ; many former players claimed that the alteration had caused a deterioration in batting and reduced the number of shots played on the off side . A 1963 report in The Times blamed the law for reducing the variety of bowling styles : \" the change has led to a steady increase in the amount of seam and off @-@ spin bowling . Whereas in the early thirties every county had a leg spinner and an orthodox left arm spinner , leg spinners , at any rate , are now few and far between . Walk on to any of the first @-@ class grounds at any time tomorrow and the chances are that you will see the wicketkeeper standing back and a medium pace bowler in action ... there is little doubt that the game , as a spectacle , is less attractive than it was . \" Several critics , including Bob Wyatt , maintained that the lbw law should be returned to its pre @-@ 1935 wording ; he campaigned to do so until his death in 1995 . On the other hand , Bradman , in the 1950s , proposed extending the law so that batsmen could be lbw even if they were struck outside the line of off stump . An MCC study of the state of cricket , carried out in 1956 and 1957 , examined the prevalent and unpopular tactic involving off @-@ spin and inswing bowlers aiming at leg stump with fielders concentrated on the leg side . Rather than alter the lbw law to combat the problem , the MCC reduced the number of fielders allowed on the leg side . \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|>According to Gerald Brodribb, in his survey and history of the Laws, the change produced<|end-PI|> <|retrieval|> <|start-KFI|>more \"enterprising\", exciting cricket but any alteration in outlook was halted by the Second World War<|end-KFI|>. <|start-PI|>When the sport resumed in 1946, batsmen were out of practice and the amended lbw law played into the hands of<|end-PI|> <|retrieval|> <|start-KFI|>off spin and inswing bowlers, who began to dominate county cricket<|end-KFI|>. <|start-PI|>The cricket historian Derek Birley notes that many of these bowlers imitated the methods of<|end-PI|> <|retrieval|> <|start-KFI|>Alec Bedser, an inswing bowler who was successful immediately after the war<|end-KFI|>, but that the resulting cricket was unexciting to watch. The revised lbw law, and other alterations in the game in favour of the bowler, further encouraged such bowling. The new law continued to provoke debate among writers and cricketers; many former players claimed that the alteration had caused a deterioration in batting and reduced the number of shots played on the off side. <|start-PI|>A 1963 report in The Times blamed the law for<|end-PI|> <|retrieval|> <|start-KFI|>reducing the variety of bowling styles<|end-KFI|>: \"the change has led to a steady increase in the amount of seam and off-spin bowling. Whereas in the early thirties every county had a leg spinner and an orthodox left arm spinner, leg spinners, at any rate, are now few and far between. Walk on to any of the first-class grounds at any time tomorrow and the chances are that you will see the wicketkeeper standing back and a medium pace bowler in action ... there is little doubt that the game, as a spectacle, is less attractive than it was.\" <|start-PI|>Several critics, including Bob Wyatt, maintained that<|end-PI|> <|retrieval|> <|start-KFI|>the lbw law should be returned to its pre-1935 wording; he campaigned to do so until his death in 1995<|end-KFI|>. <|start-PI|>On the other hand, Bradman, in the 1950s, proposed<|end-PI|> <|retrieval|> <|start-KFI|>extending the law so that batsmen could be lbw even if they were struck outside the line of off stump<|end-KFI|>. An MCC study of the state of cricket, carried out in 1956 and 1957, examined the prevalent and unpopular tactic involving off-spin and inswing bowlers aiming at leg stump with fielders concentrated on the leg side. Rather than alter the lbw law to combat the problem, the MCC reduced the number of fielders allowed on the leg side.",
"none": null,
"results": [
{
"Explanation": "The PI introduces Gerald Brodribb's perspective on the change in cricket laws, while the KFI provides the specific outcome. The <|retrieval|> token is inserted to indicate where additional information about Brodribb's survey might be needed.",
"Ideal Knowledge": "A detailed analysis of Gerald Brodribb's survey on cricket laws, focusing on the changes made and their impact on the game's style and excitement level before World War II.",
"Key Factual Information": "more \"enterprising\", exciting cricket but any alteration in outlook was halted by the Second World War",
"Prefix reformulated": "What did the change in cricket laws produce according to Gerald Brodribb?",
"Prefix-intent": "According to Gerald Brodribb, in his survey and history of the Laws, the change produced",
"Verification": "The KFI directly provides the outcome of the change as described by Brodribb, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of cricket resuming after the war and the impact of the amended lbw law, while the KFI specifies which types of bowlers benefited. The <|retrieval|> token is inserted to indicate where additional information about post-war cricket and bowling techniques might be needed.",
"Ideal Knowledge": "Information on the impact of the amended lbw law on different bowling styles in post-World War II cricket, with specific focus on off spin and inswing bowling techniques and their effectiveness in county cricket.",
"Key Factual Information": "off spin and inswing bowlers, who began to dominate county cricket",
"Prefix reformulated": "Which types of bowlers benefited from the amended lbw law when cricket resumed in 1946?",
"Prefix-intent": "When the sport resumed in 1946, batsmen were out of practice and the amended lbw law played into the hands of",
"Verification": "The KFI directly identifies the types of bowlers who benefited from the law change, as prompted by the PI."
},
{
"Explanation": "The PI introduces Derek Birley's observation about bowlers imitating a specific player, while the KFI identifies Alec Bedser and provides context. The <|retrieval|> token is inserted to indicate where additional information about Bedser and his influence might be needed.",
"Ideal Knowledge": "A biography of Alec Bedser, focusing on his bowling technique, success in post-war cricket, and his influence on other bowlers of the era.",
"Key Factual Information": "Alec Bedser, an inswing bowler who was successful immediately after the war",
"Prefix reformulated": "Whose bowling methods did many bowlers imitate according to Derek Birley?",
"Prefix-intent": "The cricket historian Derek Birley notes that many of these bowlers imitated the methods of",
"Verification": "The KFI directly identifies Alec Bedser as the bowler whose methods were imitated, as prompted by the PI."
},
{
"Explanation": "The PI introduces a report from The Times and its criticism of the law, while the KFI provides the specific effect blamed on the law. The <|retrieval|> token is inserted to indicate where additional information about the report and its findings might be needed.",
"Ideal Knowledge": "A comprehensive analysis of the 1963 report in The Times, detailing the impact of the lbw law change on bowling styles in cricket, including statistics and expert opinions.",
"Key Factual Information": "reducing the variety of bowling styles",
"Prefix reformulated": "What did a 1963 report in The Times blame the lbw law for?",
"Prefix-intent": "A 1963 report in The Times blamed the law for",
"Verification": "The KFI directly states the effect of the law change on bowling styles, as prompted by the PI."
},
{
"Explanation": "The PI introduces Bob Wyatt and other critics' stance, while the KFI provides their specific position and Wyatt's long-term commitment to it. The <|retrieval|> token is inserted to indicate where additional information about Wyatt and the debate over the lbw law might be needed.",
"Ideal Knowledge": "A detailed account of Bob Wyatt's career, his views on cricket laws, and his campaign to revert the lbw law, including arguments for and against his position.",
"Key Factual Information": "the lbw law should be returned to its pre-1935 wording; he campaigned to do so until his death in 1995",
"Prefix reformulated": "What did Bob Wyatt and other critics maintain about the lbw law?",
"Prefix-intent": "Several critics, including Bob Wyatt, maintained that",
"Verification": "The KFI directly states the position of Bob Wyatt and other critics regarding the lbw law, as prompted by the PI."
},
{
"Explanation": "The PI introduces Bradman's contrasting view, while the KFI provides the specific details of his proposal. The <|retrieval|> token is inserted to indicate where additional information about Bradman's cricket law reform ideas might be needed.",
"Ideal Knowledge": "A comprehensive analysis of Don Bradman's views on cricket laws, focusing on his 1950s proposal to extend the lbw law, including the reasoning behind his suggestion and reactions from the cricket community.",
"Key Factual Information": "extending the law so that batsmen could be lbw even if they were struck outside the line of off stump",
"Prefix reformulated": "What did Bradman propose regarding the lbw law in the 1950s?",
"Prefix-intent": "On the other hand, Bradman, in the 1950s, proposed",
"Verification": "The KFI directly states Bradman's proposal for changing the lbw law, as prompted by the PI."
}
]
} |
In the 1950s and 1960s , the amount of pad @-@ play increased , owing to more difficult and unpredictable pitches that made batting much harder . Critics continued to regard this tactic as " negative and unfair " . In an effort to discourage pad @-@ play and encourage leg spin bowling , a new variant of the lbw law was introduced , initially in Australia and the West Indies in the 1969 – 70 season , then in England for 1970 . Under the re @-@ worded law , a batsman would be lbw if a ball destined to hit the stumps pitched in line with the wickets or " outside a batsman 's off stump and in the opinion of the umpire he made no genuine attempt to play the ball with his bat . " This revision omitted the requirement that the impact should be in line with the wickets , but meant that any batsman playing a shot could not be out if the ball pitched outside off stump , in contrast to the 1935 law . The editor of Wisden believed the change encouraged batsmen to take more risks , and had produced more attractive cricket . However , the proportion of wickets falling lbw sharply declined , and concerns were expressed in Australia . The Australian authorities proposed a reversion to the previous law . A batsman could once more be out to a ball that pitched outside off stump , but a provision was added that " if no stroke is offered to a ball pitching outside the off @-@ stump which in the opinion of the umpire would hit the stumps , but hits the batsman on any part of his person other than the hand , then the batsman is out , even if that part of the person hit is not in line between wicket and wicket . " The difference to the 1935 rule was that the batsman could now be out even if the ball struck outside the line of off @-@ stump . This wording was adopted throughout the world , although it was not yet part of the official Laws , from 1972 and the percentage of lbws sharply increased to beyond the levels preceding the 1970 change . The MCC added the revised wording to the Laws of Cricket in 1980 ; this version of the lbw law is still used as of 2013 .
| [
{
"content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n In the 1950s and 1960s , the amount of pad @-@ play increased , owing to more difficult and unpredictable pitches that made batting much harder . Critics continued to regard this tactic as \" negative and unfair \" . In an effort to discourage pad @-@ play and encourage leg spin bowling , a new variant of the lbw law was introduced , initially in Australia and the West Indies in the 1969 – 70 season , then in England for 1970 . Under the re @-@ worded law , a batsman would be lbw if a ball destined to hit the stumps pitched in line with the wickets or \" outside a batsman 's off stump and in the opinion of the umpire he made no genuine attempt to play the ball with his bat . \" This revision omitted the requirement that the impact should be in line with the wickets , but meant that any batsman playing a shot could not be out if the ball pitched outside off stump , in contrast to the 1935 law . The editor of Wisden believed the change encouraged batsmen to take more risks , and had produced more attractive cricket . However , the proportion of wickets falling lbw sharply declined , and concerns were expressed in Australia . The Australian authorities proposed a reversion to the previous law . A batsman could once more be out to a ball that pitched outside off stump , but a provision was added that \" if no stroke is offered to a ball pitching outside the off @-@ stump which in the opinion of the umpire would hit the stumps , but hits the batsman on any part of his person other than the hand , then the batsman is out , even if that part of the person hit is not in line between wicket and wicket . \" The difference to the 1935 rule was that the batsman could now be out even if the ball struck outside the line of off @-@ stump . This wording was adopted throughout the world , although it was not yet part of the official Laws , from 1972 and the percentage of lbws sharply increased to beyond the levels preceding the 1970 change . The MCC added the revised wording to the Laws of Cricket in 1980 ; this version of the lbw law is still used as of 2013 . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.",
"role": "user"
}
] | {
"Updated_Passage": "In the 1950s and 1960s , <|start-PI|> the amount of pad @-@ play increased , owing to <|end-PI|> <|retrieval|> <|start-KFI|> more difficult and unpredictable pitches that made batting much harder <|end-KFI|> . Critics continued to regard this tactic as \" negative and unfair \" . <|start-PI|> In an effort to discourage pad @-@ play and encourage leg spin bowling , a new variant of the lbw law was introduced , initially in <|end-PI|> <|retrieval|> <|start-KFI|> Australia and the West Indies in the 1969 – 70 season , then in England for 1970 <|end-KFI|> . <|start-PI|> Under the re @-@ worded law , a batsman would be lbw if <|end-PI|> <|retrieval|> <|start-KFI|> a ball destined to hit the stumps pitched in line with the wickets or \" outside a batsman 's off stump and in the opinion of the umpire he made no genuine attempt to play the ball with his bat . \" <|end-KFI|> This revision omitted the requirement that the impact should be in line with the wickets , but meant that any batsman playing a shot could not be out if the ball pitched outside off stump , in contrast to the 1935 law . <|start-PI|> The editor of Wisden believed the change encouraged batsmen to take more risks , and had produced <|end-PI|> <|retrieval|> <|start-KFI|> more attractive cricket <|end-KFI|> . However , the proportion of wickets falling lbw sharply declined , and concerns were expressed in Australia . The Australian authorities proposed a reversion to the previous law . A batsman could once more be out to a ball that pitched outside off stump , but a provision was added that \" if no stroke is offered to a ball pitching outside the off @-@ stump which in the opinion of the umpire would hit the stumps , but hits the batsman on any part of his person other than the hand , then the batsman is out , even if that part of the person hit is not in line between wicket and wicket . \" The difference to the 1935 rule was that the batsman could now be out even if the ball struck outside the line of off @-@ stump . This wording was adopted throughout the world , although it was not yet part of the official Laws , from 1972 and the percentage of lbws sharply increased to beyond the levels preceding the 1970 change . <|start-PI|> The MCC added the revised wording to the Laws of Cricket in <|end-PI|> <|retrieval|> <|start-KFI|> 1980 ; this version of the lbw law is still used as of 2013 <|end-KFI|> .",
"none": null,
"results": [
{
"Explanation": "The PI sets up the context of increased pad-play in cricket, while the KFI provides the specific reason for this change. The <|retrieval|> token is inserted to indicate where additional information about pitch conditions might be needed.",
"Ideal Knowledge": "Information about cricket pitch conditions in the 1950s and 1960s, explaining how they became more challenging for batsmen and led to increased defensive tactics like pad-play.",
"Key Factual Information": "more difficult and unpredictable pitches that made batting much harder",
"Prefix reformulated": "What caused the increase in pad-play in cricket during the 1950s and 1960s?",
"Prefix-intent": "In the 1950s and 1960s , the amount of pad @-@ play increased , owing to",
"Verification": "The KFI directly provides the reason for the increase in pad-play as requested in the PI."
},
{
"Explanation": "The PI introduces the context of a new lbw law variant, and the KFI provides the specific locations and times of its introduction. The <|retrieval|> token is inserted to indicate where additional information about cricket rule changes might be needed.",
"Ideal Knowledge": "A chronological account of changes to the lbw (leg before wicket) law in cricket, specifically detailing the introduction of new variants in different countries during the 1969-70 season.",
"Key Factual Information": "Australia and the West Indies in the 1969 – 70 season , then in England for 1970",
"Prefix reformulated": "Where and when was a new variant of the lbw law introduced to discourage pad-play and encourage leg spin bowling in cricket?",
"Prefix-intent": "In an effort to discourage pad @-@ play and encourage leg spin bowling , a new variant of the lbw law was introduced , initially in",
"Verification": "The KFI directly provides the locations and timeframes for the introduction of the new lbw law variant, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of the reworded lbw law, and the KFI provides the specific conditions for an lbw decision. The <|retrieval|> token is inserted to indicate where additional information about cricket rules might be needed.",
"Ideal Knowledge": "Detailed explanation of the reworded lbw law in cricket, including specific conditions for a batsman to be given out, with emphasis on ball trajectory and batsman's actions.",
"Key Factual Information": "a ball destined to hit the stumps pitched in line with the wickets or \" outside a batsman 's off stump and in the opinion of the umpire he made no genuine attempt to play the ball with his bat . \"",
"Prefix reformulated": "What were the conditions for a batsman to be given out lbw under the reworded cricket law?",
"Prefix-intent": "Under the re @-@ worded law , a batsman would be lbw if",
"Verification": "The KFI directly provides the conditions for an lbw decision under the new law, as prompted by the PI."
},
{
"Explanation": "The PI introduces the Wisden editor's belief about the law change, and the KFI provides the specific outcome he believed it produced. The <|retrieval|> token is inserted to indicate where additional expert opinions might be needed.",
"Ideal Knowledge": "Analysis from cricket experts, particularly the editor of Wisden, on the impact of the lbw law change on batting strategies and overall game attractiveness.",
"Key Factual Information": "more attractive cricket",
"Prefix reformulated": "What did the editor of Wisden believe was the result of the change in the lbw law?",
"Prefix-intent": "The editor of Wisden believed the change encouraged batsmen to take more risks , and had produced",
"Verification": "The KFI directly provides the editor's opinion on the result of the law change, as prompted by the PI."
},
{
"Explanation": "The PI sets up the context of the MCC adding the revised lbw law, and the KFI provides the specific year and its continued use. The <|retrieval|> token is inserted to indicate where additional information about cricket law updates might be needed.",
"Ideal Knowledge": "A timeline of changes to the Laws of Cricket, specifically focusing on the lbw law revisions, their adoption by the MCC, and their longevity in the game.",
"Key Factual Information": "1980 ; this version of the lbw law is still used as of 2013",
"Prefix reformulated": "When did the MCC add the revised wording of the lbw law to the Laws of Cricket?",
"Prefix-intent": "The MCC added the revised wording to the Laws of Cricket in",
"Verification": "The KFI directly provides the year when the MCC added the revised wording and its continued use, as prompted by the PI."
}
]
} |