Datasets:

Self-GRIT

/

wikitext-2-raw-v1-preprocessed-5k-PI_KFI_claude-claude-3-5-sonnet-20240620

like 2

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 an estimated 27 @,@ 000 civil aircraft registered in the UK , 96 per cent of which are engaged in GA activities . In 2005 the GA fleet comprised 9 @,@ 000 fixed @-@ wing aircraft , 4 @,@ 100 microlights , 1 @,@ 300 helicopters , 1 @,@ 800 airships / balloons , 2 @,@ 500 gliders and some 7 @,@ 000 hang gliders . Estimates put the number of foreign @-@ registered GA aircraft based in the UK at 900 . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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|>There are an estimated<|end-PI|> <|retrieval|> <|start-KFI|>27,000 civil aircraft registered in the UK<|end-KFI|>, <|start-PI|>96 per cent of which are engaged in<|end-PI|> <|retrieval|> <|start-KFI|>GA activities<|end-KFI|>. <|start-PI|>In 2005 the GA fleet comprised<|end-PI|> <|retrieval|> <|start-KFI|>9,000 fixed-wing aircraft, 4,100 microlights, 1,300 helicopters, 1,800 airships/balloons, 2,500 gliders and some 7,000 hang gliders<|end-KFI|>. <|start-PI|>Estimates put the number of foreign-registered GA aircraft based in the UK at<|end-PI|> <|retrieval|> <|start-KFI|>900<|end-KFI|>.", "none": null, "results": [ { "Explanation": "The PI sets up a query about the number of registered civil aircraft, and the KFI provides the specific figure. The <|retrieval|> token is inserted to indicate where additional information might be needed to support this factual statement.", "Ideal Knowledge": "A comprehensive database of civil aircraft registrations in the UK, including historical data and current statistics.", "Key Factual Information": "27,000 civil aircraft registered in the UK", "Prefix reformulated": "How many civil aircraft are registered in the UK?", "Prefix-intent": "There are an estimated", "Verification": "The KFI directly provides the number of registered civil aircraft in the UK, as prompted by the PI." }, { "Explanation": "The PI introduces a percentage and sets up a query about the type of activities, while the KFI specifies that these are GA activities. The <|retrieval|> token is inserted to indicate where additional information about GA activities might be needed.", "Ideal Knowledge": "A breakdown of civil aircraft usage in the UK, detailing the percentage of aircraft involved in general aviation (GA) activities versus other types of operations.", "Key Factual Information": "GA activities", "Prefix reformulated": "What percentage of UK registered civil aircraft are engaged in GA activities?", "Prefix-intent": "96 per cent of which are engaged in", "Verification": "The KFI directly states the type of activities that 96% of the registered aircraft are engaged in, as prompted by the PI." }, { "Explanation": "The PI sets up a query about the composition of the GA fleet in a specific year, and the KFI provides a comprehensive breakdown. The <|retrieval|> token is inserted to indicate where additional information about the fleet composition might be needed.", "Ideal Knowledge": "A detailed inventory of the UK's general aviation fleet in 2005, categorized by aircraft type with specific numbers for each category.", "Key Factual Information": "9,000 fixed-wing aircraft, 4,100 microlights, 1,300 helicopters, 1,800 airships/balloons, 2,500 gliders and some 7,000 hang gliders", "Prefix reformulated": "What was the composition of the UK GA fleet in 2005?", "Prefix-intent": "In 2005 the GA fleet comprised", "Verification": "The KFI directly provides the detailed composition of the GA fleet in 2005, as prompted by the PI." }, { "Explanation": "The PI sets up a query about the number of foreign-registered GA aircraft in the UK, and the KFI provides the specific estimate. The <|retrieval|> token is inserted to indicate where additional information about foreign-registered aircraft might be needed.", "Ideal Knowledge": "Data on foreign-registered aircraft based in the UK, including historical trends and current estimates.", "Key Factual Information": "900", "Prefix reformulated": "How many foreign-registered GA aircraft are estimated to be based in the UK?", "Prefix-intent": "Estimates put the number of foreign-registered GA aircraft based in the UK at", "Verification": "The KFI directly provides the estimated number of foreign-registered GA aircraft in the UK, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 number of pilots licensed by the CAA to fly powered aircraft in 2005 was 47 @,@ 000 , of whom 28 @,@ 000 held a Private Pilot Licence . The remainder held professional pilot licences , either a Commercial Pilot Licence or an Airline Transport Pilot Licence , although not all of these would be engaged in GA activities . In addition , there are 10 @,@ 000 active glider pilots , and estimates put the membership of aviation @-@ related sport and recreational associations at 36 @,@ 000 . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.", "role": "user" } ]

{ "Updated_Passage": null, "none": "", "results": null }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 number of aerodromes that support GA in the UK is difficult to establish with certainty . Pooleys 2008 United Kingdom Flight Guide lists 355 , and the Airplan Flight Equipment UK VFR Flight Guide 2008 lists nearly 500 . Lockyears Farm ' Strips ' and Private Airfields Flight Guide lists more than 300 landing sites . The GASAR study estimates 1 @,@ 100 formal flying sites in England alone , a figure which includes 400 sites known to planning authorities but not included in flight guides . It estimates another 759 informal sites known only to land owners , customs , and members of the enthusiast group Air @-@ Britain . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 number of aerodromes that support GA in the UK is difficult to establish with certainty<|end-PI|> <|retrieval|> <|start-KFI|>Pooleys 2008 United Kingdom Flight Guide lists 355 , and the Airplan Flight Equipment UK VFR Flight Guide 2008 lists nearly 500<|end-KFI|> . <|start-PI|>Lockyears Farm ' Strips ' and Private Airfields Flight Guide lists<|end-PI|> <|retrieval|> <|start-KFI|>more than 300 landing sites<|end-KFI|> . <|start-PI|>The GASAR study estimates<|end-PI|> <|retrieval|> <|start-KFI|>1 @,@ 100 formal flying sites in England alone , a figure which includes 400 sites known to planning authorities but not included in flight guides<|end-KFI|> . <|start-PI|>It estimates another<|end-PI|> <|retrieval|> <|start-KFI|>759 informal sites known only to land owners , customs , and members of the enthusiast group Air @-@ Britain<|end-KFI|> .", "none": null, "results": [ { "Explanation": "The PI introduces the difficulty in determining the exact number of GA aerodromes in the UK, while the KFI provides specific numbers from different sources. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.", "Ideal Knowledge": "A comprehensive list of all aerodromes supporting General Aviation in the UK, including those listed in various flight guides and those known to local authorities but not officially listed.", "Key Factual Information": "Pooleys 2008 United Kingdom Flight Guide lists 355 , and the Airplan Flight Equipment UK VFR Flight Guide 2008 lists nearly 500", "Prefix reformulated": "How many aerodromes support General Aviation (GA) in the UK?", "Prefix-intent": "The number of aerodromes that support GA in the UK is difficult to establish with certainty", "Verification": "The KFI directly provides information about the number of aerodromes listed in different flight guides, addressing the uncertainty mentioned in the PI." }, { "Explanation": "The PI introduces a specific flight guide, and the KFI provides the number of landing sites listed in that guide. The <|retrieval|> token is inserted to indicate where additional information about this specific guide might be needed.", "Ideal Knowledge": "Detailed information from the Lockyears Farm 'Strips' and Private Airfields Flight Guide, including a complete list of all landing sites and their characteristics.", "Key Factual Information": "more than 300 landing sites", "Prefix reformulated": "How many landing sites does Lockyears Farm 'Strips' and Private Airfields Flight Guide list?", "Prefix-intent": "Lockyears Farm ' Strips ' and Private Airfields Flight Guide lists", "Verification": "The KFI directly provides the number of landing sites listed in the specific guide mentioned in the PI." }, { "Explanation": "The PI introduces the GASAR study, and the KFI provides specific estimates and details from the study. The <|retrieval|> token is inserted to indicate where additional information about the GASAR study and its findings might be needed.", "Ideal Knowledge": "The full GASAR (General Aviation Strategic Action Report) study, including its methodology, data sources, and detailed breakdown of flying sites in England and their classification.", "Key Factual Information": "1 @,@ 100 formal flying sites in England alone , a figure which includes 400 sites known to planning authorities but not included in flight guides", "Prefix reformulated": "What does the GASAR study estimate about flying sites in England?", "Prefix-intent": "The GASAR study estimates", "Verification": "The KFI directly provides the estimate from the GASAR study, including specific numbers and details about the sites, as prompted by the PI." }, { "Explanation": "The PI continues from the previous GASAR study mention, and the KFI provides specific information about informal sites. The <|retrieval|> token is inserted to indicate where additional information about these informal sites might be needed.", "Ideal Knowledge": "Detailed information from the GASAR study about informal flying sites, including their distribution, characteristics, and the methodology used to estimate their number.", "Key Factual Information": "759 informal sites known only to land owners , customs , and members of the enthusiast group Air @-@ Britain", "Prefix reformulated": "How many informal sites does the GASAR study estimate?", "Prefix-intent": "It estimates another", "Verification": "The KFI directly provides the number of informal sites estimated by the GASAR study, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 sector was estimated to employ nearly 12 @,@ 000 people and directly contribute £ 1 @.@ 4 billion to the UK economy in 2005 , making it roughly seven per cent of the size of the CAT industry . Nearly half of the economic contribution was generated by business aviation . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Most sectors of GA for which data are available have experienced growth in aircraft numbers and hours flown over the last two decades . The lighter end of the GA spectrum : microlights , amateur built , and airships and balloons , have in particular shown strong growth , although the last of these activities was severely curtailed during the foot @-@ and @-@ mouth outbreak in 2001 , when access to farmland was denied . After strong growth in the late 1980s , traditional flying has shown a slight decline recently , reflecting a move amongst recreational flyers towards microlight aircraft , and increased numbers of foreign @-@ registered aircraft . Recreational helicopter usage has grown primarily due to the introduction of smaller and cheaper aircraft . Glider activity has remained relatively static , although there has been a gradual increase in the number of self @-@ launching motor gliders . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Business aviation has shown strong growth , although the numbers of aircraft on the UK register have declined . This reflects a shift away from turboprop aircraft towards foreign @-@ registered business jets based in the UK , which are estimated to be growing in numbers . However , twin piston @-@ engined aircraft numbers have declined significantly , reflecting pressures on the light air @-@ taxi segment from increasingly flexible and cheaper scheduled services , and a more sophisticated corporate charter business . The amount of flight training conducted by UK schools has declined , largely at the hands of competition from foreign schools , which benefit from lower costs and better weather . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Since 1990 the total number of hours flown annually by the GA sector has remained in the range 1 @.@ 25 – 1 @.@ 35 million , the dominant sector being traditional GA flying , which accounts for 0 @.@ 6 million per year . An overall increase in aircraft numbers combined with nil growth in hours flown has brought the annual average utilisation per aircraft down from 157 hours in 1984 to 103 hours in 2002 . The decline in asset utilisation has led to speculation that the economic health of the GA industry is weakening , though the lack of data on profitability makes this difficult to confirm . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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|>Since 1990 the total number of hours flown annually by the GA sector has remained in the range<|end-PI|> <|retrieval|> <|start-KFI|>1 @.@ 25 – 1 @.@ 35 million<|end-KFI|>, <|start-PI|>the dominant sector being traditional GA flying , which accounts for<|end-PI|> <|retrieval|> <|start-KFI|>0 @.@ 6 million per year<|end-KFI|>. <|start-PI|>An overall increase in aircraft numbers combined with nil growth in hours flown has brought the annual average utilisation per aircraft down from 157 hours in 1984 to<|end-PI|> <|retrieval|> <|start-KFI|>103 hours in 2002<|end-KFI|>. The decline in asset utilisation has led to speculation that the economic health of the GA industry is weakening , though the lack of data on profitability makes this difficult to confirm.", "none": null, "results": [ { "Explanation": "The PI sets up the context for the annual flight hours of the GA sector, and the KFI provides the specific range. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.", "Ideal Knowledge": "A comprehensive dataset of annual flight hours for the General Aviation (GA) sector from 1990 onwards, showing the consistent range between 1.25 and 1.35 million hours.", "Key Factual Information": "1 @.@ 25 – 1 @.@ 35 million", "Prefix reformulated": "What is the range of total hours flown annually by the GA sector since 1990?", "Prefix-intent": "Since 1990 the total number of hours flown annually by the GA sector has remained in the range", "Verification": "The KFI directly provides the range of hours flown, as requested in the PI." }, { "Explanation": "The PI identifies traditional GA flying as the dominant sector and sets up a query about its contribution, while the KFI provides the specific annual hours. The <|retrieval|> token is inserted to indicate where additional information might be needed to support this fact.", "Ideal Knowledge": "A breakdown of flight hours within the GA sector, specifically highlighting that traditional GA flying accounts for 0.6 million hours annually.", "Key Factual Information": "0 @.@ 6 million per year", "Prefix reformulated": "How many hours does traditional GA flying account for annually?", "Prefix-intent": "the dominant sector being traditional GA flying , which accounts for", "Verification": "The KFI directly provides the annual hours for traditional GA flying, as prompted by the PI." }, { "Explanation": "The PI provides context about the change in aircraft utilization over time, and the KFI gives the specific figure for 2002. The <|retrieval|> token is inserted to indicate where additional information might be needed to support this trend.", "Ideal Knowledge": "Historical data on GA aircraft utilization, showing the decline from 157 hours per aircraft in 1984 to 103 hours in 2002, along with information on the increase in aircraft numbers and stagnant total flight hours.", "Key Factual Information": "103 hours in 2002", "Prefix reformulated": "What was the annual average utilisation per aircraft in 2002?", "Prefix-intent": "An overall increase in aircraft numbers combined with nil growth in hours flown has brought the annual average utilisation per aircraft down from 157 hours in 1984 to", "Verification": "The KFI directly provides the annual average utilisation per aircraft in 2002, as set up by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 objective of regulation is to \" promote high standards of safety in all aspects of aviation \" , and this is the main area of interaction between the CAA and the GA sector . Efforts focus on assuring appropriate standards of airworthiness , pilot qualification , the rules for the movement of aircraft , and equipment to be carried . The CAA was established as the primary regulatory body for all aviation in the UK in 1972 . In 1991 it started working within the Joint Aviation Authorities ( JAA ) framework to implement agreed common standards , known as the Joint Aviation Requirements ( JAR ) , throughout the European Union ( EU ) . In 2003 this was taken a step further when the European Aviation Safety Agency ( EASA ) was established as the central EU regulator , taking over responsibility for legislating airworthiness and environmental regulation from the national authorities . The CAA acts as an agency of EASA on these issues , retaining its original regulatory powers in areas not yet transferred to EASA . Proposed developments seek to establish EASA as the single authority throughout the EU , taking over from individual member states the power to regulate all aviation other than that specifically excluded from the scope of EASA . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Within this framework certain sectors of GA are governed on a devolved basis . In all cases the CAA / EASA retains responsibility for safety regulation , but representative bodies , particularly of sectors that are not included in the scope of EASA , are granted greater oversight of their activities . The majority of microlight aircraft are regulated by the British Microlight Aircraft Association ( BMAA ) , although a significant number are regulated by the Light Aircraft Association ( LAA ) , formerly known as the Popular Flying Association . The LAA is the primary regulator for amateur built aircraft , as well as vintage and classic aircraft . Parachuting is governed by the British Parachute Association , although the aircraft used in this activity are generally CAA @-@ regulated . Balloon and airship flying is overseen by the British Balloon and Airship Club . The UK @-@ specific National Private Pilot Licence ( NPPL ) is administered by the National Pilots Licensing Group Ltd . , supported by the LAA , the Aircraft Owners and Pilots Association UK , the British Gliding Association , and the British Microlight Aircraft Association . Separate from these devolved groups , gliding in the UK is self @-@ regulated . The British Gliding Association was until recently responsible for glider airworthiness , now formally regulated as a result of EASA legislation , and still retains control of pilot certification . Hang gliding and paragliding activities ( i.e. foot @-@ launched gliders ) are governed by the British Hang Gliding and Paragliding Association . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.", "role": "user" } ]

{ "Updated_Passage": "Within this framework certain sectors of GA are governed on a devolved basis . In all cases the CAA / EASA retains responsibility for safety regulation , but representative bodies , particularly of sectors that are not included in the scope of EASA , are granted greater oversight of their activities . <|start-PI|> The majority of microlight aircraft are regulated by <|end-PI|> <|retrieval|> <|start-KFI|> the British Microlight Aircraft Association ( BMAA ) <|end-KFI|> , although a significant number are regulated by the Light Aircraft Association ( LAA ) , formerly known as the Popular Flying Association . <|start-PI|> The LAA is the primary regulator for <|end-PI|> <|retrieval|> <|start-KFI|> amateur built aircraft , as well as vintage and classic aircraft <|end-KFI|> . <|start-PI|> Parachuting is governed by <|end-PI|> <|retrieval|> <|start-KFI|> the British Parachute Association <|end-KFI|> , although the aircraft used in this activity are generally CAA @-@ regulated . <|start-PI|> Balloon and airship flying is overseen by <|end-PI|> <|retrieval|> <|start-KFI|> the British Balloon and Airship Club <|end-KFI|> . <|start-PI|> The UK @-@ specific National Private Pilot Licence ( NPPL ) is administered by <|end-PI|> <|retrieval|> <|start-KFI|> the National Pilots Licensing Group Ltd . , supported by the LAA , the Aircraft Owners and Pilots Association UK , the British Gliding Association , and the British Microlight Aircraft Association <|end-KFI|> . Separate from these devolved groups , gliding in the UK is self @-@ regulated . The British Gliding Association was until recently responsible for glider airworthiness , now formally regulated as a result of EASA legislation , and still retains control of pilot certification . <|start-PI|> Hang gliding and paragliding activities ( i.e. foot @-@ launched gliders ) are governed by <|end-PI|> <|retrieval|> <|start-KFI|> the British Hang Gliding and Paragliding Association <|end-KFI|> .", "none": null, "results": [ { "Explanation": "The PI sets up the context for microlight aircraft regulation, and the KFI provides the specific organization responsible. The <|retrieval|> token is inserted to indicate where additional information about regulatory bodies might be needed.", "Ideal Knowledge": "A comprehensive list of aviation regulatory bodies in the UK, detailing their responsibilities and the types of aircraft they oversee, with specific emphasis on the BMAA's role in microlight regulation.", "Key Factual Information": "the British Microlight Aircraft Association ( BMAA )", "Prefix reformulated": "Which organization regulates the majority of microlight aircraft?", "Prefix-intent": "The majority of microlight aircraft are regulated by", "Verification": "The KFI directly provides the organization that regulates the majority of microlight aircraft, as prompted by the PI." }, { "Explanation": "The PI introduces the LAA's regulatory role, and the KFI specifies the types of aircraft it oversees. The <|retrieval|> token is inserted to indicate where additional information about aircraft categories might be needed.", "Ideal Knowledge": "Detailed information about the Light Aircraft Association's regulatory responsibilities, including definitions and examples of amateur-built, vintage, and classic aircraft categories.", "Key Factual Information": "amateur built aircraft , as well as vintage and classic aircraft", "Prefix reformulated": "What types of aircraft does the Light Aircraft Association (LAA) primarily regulate?", "Prefix-intent": "The LAA is the primary regulator for", "Verification": "The KFI directly lists the types of aircraft regulated by the LAA, as prompted by the PI." }, { "Explanation": "The PI sets up the context for parachuting governance, and the KFI provides the specific organization responsible. The <|retrieval|> token is inserted to indicate where additional information about parachuting regulation might be needed.", "Ideal Knowledge": "Information about the British Parachute Association, its role in regulating parachuting activities, and its relationship with other aviation regulatory bodies in the UK.", "Key Factual Information": "the British Parachute Association", "Prefix reformulated": "Which organization governs parachuting in the UK?", "Prefix-intent": "Parachuting is governed by", "Verification": "The KFI directly provides the organization that governs parachuting, as prompted by the PI." }, { "Explanation": "The PI sets up the context for balloon and airship regulation, and the KFI provides the specific organization responsible. The <|retrieval|> token is inserted to indicate where additional information about balloon and airship regulation might be needed.", "Ideal Knowledge": "Information about the British Balloon and Airship Club, its responsibilities in overseeing balloon and airship activities, and its relationship with other aviation regulatory bodies in the UK.", "Key Factual Information": "the British Balloon and Airship Club", "Prefix reformulated": "Which organization oversees balloon and airship flying in the UK?", "Prefix-intent": "Balloon and airship flying is overseen by", "Verification": "The KFI directly provides the organization that oversees balloon and airship flying, as prompted by the PI." }, { "Explanation": "The PI introduces the NPPL and sets up a query about its administration, while the KFI provides the specific organization and its supporting bodies. The <|retrieval|> token is inserted to indicate where additional information about pilot licensing might be needed.", "Ideal Knowledge": "Detailed information about the National Pilots Licensing Group Ltd., its role in administering the NPPL, and the specific contributions of each supporting organization mentioned.", "Key Factual Information": "the National Pilots Licensing Group Ltd . , supported by the LAA , the Aircraft Owners and Pilots Association UK , the British Gliding Association , and the British Microlight Aircraft Association", "Prefix reformulated": "Which organization administers the UK-specific National Private Pilot Licence (NPPL)?", "Prefix-intent": "The UK @-@ specific National Private Pilot Licence ( NPPL ) is administered by", "Verification": "The KFI directly provides the organization that administers the NPPL and its supporting organizations, as prompted by the PI." }, { "Explanation": "The PI sets up the context for hang gliding and paragliding governance, and the KFI provides the specific organization responsible. The <|retrieval|> token is inserted to indicate where additional information about hang gliding and paragliding regulation might be needed.", "Ideal Knowledge": "Information about the British Hang Gliding and Paragliding Association, its role in regulating foot-launched glider activities, and its relationship with other aviation regulatory bodies in the UK.", "Key Factual Information": "the British Hang Gliding and Paragliding Association", "Prefix reformulated": "Which organization governs hang gliding and paragliding activities in the UK?", "Prefix-intent": "Hang gliding and paragliding activities ( i.e. foot @-@ launched gliders ) are governed by", "Verification": "The KFI directly provides the organization that governs hang gliding and paragliding activities, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Under CAA and EASA rules , all aircraft are required to meet certain standards of airworthiness to fly safely and legally . Aircraft that meet these standards are issued with a Certificate of Airworthiness . However , British @-@ registered aircraft which are excluded from the scope of EASA , and which cannot satisfy the requirements for the issue of a Certificate of Airworthiness , may be issued with a Permit to Fly . This allows them to fly in UK airspace subject to certain limitations , for example being restricted to day @-@ time flights under visual flight rules only . A number of organisations ( e.g. the British Microlight Aircraft Association and the Light Aircraft Association ) have obtained a standing over @-@ flight permission for Permit to Fly aircraft within their area of interest with some European countries , notably France . Permits are typically issued to vintage and historic aircraft , amateur built aircraft , and microlights . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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|>Under CAA and EASA rules, all aircraft are required to meet certain standards of airworthiness<|end-PI|> <|retrieval|> <|start-KFI|>to fly safely and legally<|end-KFI|>. <|start-PI|>Aircraft that meet these standards are issued with<|end-PI|> <|retrieval|> <|start-KFI|>a Certificate of Airworthiness<|end-KFI|>. However, <|start-PI|>British-registered aircraft which are excluded from the scope of EASA, and which cannot satisfy the requirements for the issue of a Certificate of Airworthiness, may be issued with<|end-PI|> <|retrieval|> <|start-KFI|>a Permit to Fly<|end-KFI|>. <|start-PI|>This allows them to fly in UK airspace subject to certain limitations, for example<|end-PI|> <|retrieval|> <|start-KFI|>being restricted to day-time flights under visual flight rules only<|end-KFI|>. A number of organisations (e.g. the British Microlight Aircraft Association and the Light Aircraft Association) have obtained a standing over-flight permission for Permit to Fly aircraft within their area of interest with some European countries, notably France. <|start-PI|>Permits are typically issued to<|end-PI|> <|retrieval|> <|start-KFI|>vintage and historic aircraft, amateur built aircraft, and microlights<|end-KFI|>.", "none": null, "results": [ { "Explanation": "The PI introduces the regulatory bodies and sets up a query about aircraft requirements, while the KFI provides the purpose of these requirements. The <|retrieval|> token is inserted to indicate where additional information about specific standards might be needed.", "Ideal Knowledge": "Detailed information about CAA (Civil Aviation Authority) and EASA (European Union Aviation Safety Agency) regulations, specifically focusing on airworthiness standards and their importance for safe and legal flight operations.", "Key Factual Information": "to fly safely and legally", "Prefix reformulated": "What are aircraft required to meet under CAA and EASA rules?", "Prefix-intent": "Under CAA and EASA rules, all aircraft are required to meet certain standards of airworthiness", "Verification": "The KFI directly provides the purpose of meeting the standards mentioned in the PI." }, { "Explanation": "The PI sets up a query about what is issued to compliant aircraft, and the KFI provides the specific document. The <|retrieval|> token is inserted to indicate where additional information about the certificate might be needed.", "Ideal Knowledge": "Information about the Certificate of Airworthiness, including its purpose, issuance process, and significance in aviation safety and regulation.", "Key Factual Information": "a Certificate of Airworthiness", "Prefix reformulated": "What are aircraft that meet CAA and EASA standards issued with?", "Prefix-intent": "Aircraft that meet these standards are issued with", "Verification": "The KFI directly provides the document issued to aircraft meeting the standards mentioned in the PI." }, { "Explanation": "The PI describes a specific category of aircraft and sets up a query about an alternative document, while the KFI provides the name of this document. The <|retrieval|> token is inserted to indicate where additional information about the Permit to Fly might be needed.", "Ideal Knowledge": "Detailed information about the Permit to Fly, including its purpose, issuance criteria, and how it differs from a Certificate of Airworthiness, specifically for British-registered aircraft outside EASA scope.", "Key Factual Information": "a Permit to Fly", "Prefix reformulated": "What may be issued to British-registered aircraft excluded from EASA scope and unable to meet Certificate of Airworthiness requirements?", "Prefix-intent": "British-registered aircraft which are excluded from the scope of EASA, and which cannot satisfy the requirements for the issue of a Certificate of Airworthiness, may be issued with", "Verification": "The KFI directly provides the alternative document issued to aircraft that don't meet the standard requirements, as described in the PI." }, { "Explanation": "The PI introduces the concept of limitations for Permit to Fly aircraft, and the KFI provides specific examples. The <|retrieval|> token is inserted to indicate where additional information about these limitations might be needed.", "Ideal Knowledge": "Comprehensive list of limitations and restrictions for aircraft operating under a Permit to Fly in UK airspace, including details about visual flight rules and day-time flight requirements.", "Key Factual Information": "being restricted to day-time flights under visual flight rules only", "Prefix reformulated": "What are the limitations for aircraft with a Permit to Fly in UK airspace?", "Prefix-intent": "This allows them to fly in UK airspace subject to certain limitations, for example", "Verification": "The KFI directly provides specific examples of limitations mentioned in the PI." }, { "Explanation": "The PI sets up a query about the types of aircraft that typically receive Permits to Fly, and the KFI provides a list of these aircraft types. The <|retrieval|> token is inserted to indicate where additional information about these aircraft categories might be needed.", "Ideal Knowledge": "Detailed categorization of aircraft types that commonly operate under Permits to Fly, including specific characteristics of vintage, historic, amateur-built, and microlight aircraft.", "Key Factual Information": "vintage and historic aircraft, amateur built aircraft, and microlights", "Prefix reformulated": "What types of aircraft are typically issued Permits to Fly?", "Prefix-intent": "Permits are typically issued to", "Verification": "The KFI directly lists the types of aircraft that typically receive Permits to Fly, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 pilot qualification most relevant to GA is the Private Pilot Licence ( PPL ) , which permits the holder to fly for recreational purposes without remuneration . In addition to the European @-@ wide Joint Aviation Regulations Flight Crew Licensing ( JAR @-@ FCL ) standard , the CAA also issues UK @-@ specific national licences . In the absence of European standards for gyroplane , balloon , and airship pilots , the CAA licenses these according to the original UK PPL standard . As a response to the perception that JAR pilot licensing standards are excessively bureaucratic and expensive for the purposes of recreational pilots , the National Private Pilot Licence ( NPPL ) was introduced in 2002 . The NPPL is easier to obtain than the JAR @-@ FCL licence , has less stringent medical requirements , is more restrictive in the privileges it grants , and is valid only for flights in British @-@ registered aircraft flying in UK and French airspace . Although there are plans to bring glider pilot licensing within the regulatory framework of EASA , the gliding sector is currently self @-@ regulating in this respect . The British Gliding Association is responsible for defining the standards of initial training , and certifying , via a badge system , pilots who meet those standards . Pilots working in sectors of GA that are commercial operations , such as aerial work and business aviation , are required to hold a professional pilot licence which , at a minimum , is the Commercial Pilot Licence . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 1995 and 2004 there were 2 @,@ 630 accidents involving GA aircraft , of which 139 were fatal , resulting in the loss of 317 lives . The majority of accidents involved small fixed @-@ wing aircraft engaged in private flights , and analysis attributes the most common causes of these to : flight handling skills ; poor judgement or airmanship ; lack of training or experience ; and omission of , or inappropriate , action . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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|>Between 1995 and 2004 there were<|end-PI|> <|retrieval|> <|start-KFI|>2,630 accidents involving GA aircraft, of which 139 were fatal, resulting in the loss of 317 lives<|end-KFI|>. <|start-PI|>The majority of accidents involved<|end-PI|> <|retrieval|> <|start-KFI|>small fixed-wing aircraft engaged in private flights<|end-KFI|>, and <|start-PI|>analysis attributes the most common causes of these to<|end-PI|> <|retrieval|> <|start-KFI|>flight handling skills; poor judgement or airmanship; lack of training or experience; and omission of, or inappropriate, action<|end-KFI|>.", "none": null, "results": [ { "Explanation": "The PI sets up a query about the number of accidents in a specific timeframe, while the KFI provides the detailed statistics. The <|retrieval|> token is inserted to indicate where additional information might be needed to support these facts.", "Ideal Knowledge": "A comprehensive report on General Aviation (GA) accidents between 1995 and 2004, including total number of accidents, fatal accidents, and resulting casualties.", "Key Factual Information": "2,630 accidents involving GA aircraft, of which 139 were fatal, resulting in the loss of 317 lives", "Prefix reformulated": "How many accidents involving GA aircraft occurred between 1995 and 2004?", "Prefix-intent": "Between 1995 and 2004 there were", "Verification": "The KFI directly provides the number of accidents and additional details related to the timeframe specified in the PI." }, { "Explanation": "The PI introduces a query about the most common type of aircraft involved in accidents, and the KFI provides the specific information. The <|retrieval|> token is inserted to indicate where additional details about aircraft types might be needed.", "Ideal Knowledge": "A breakdown of GA accidents by aircraft type and flight purpose, highlighting the prevalence of small fixed-wing aircraft in private flights.", "Key Factual Information": "small fixed-wing aircraft engaged in private flights", "Prefix reformulated": "What type of aircraft were involved in the majority of GA accidents?", "Prefix-intent": "The majority of accidents involved", "Verification": "The KFI directly specifies the type of aircraft and flight category most commonly involved in accidents, as prompted by the PI." }, { "Explanation": "The PI sets up a query about the most common causes of accidents, and the KFI provides a list of these causes. The <|retrieval|> token is inserted to indicate where additional information about accident causes and their analysis might be needed.", "Ideal Knowledge": "A detailed analysis of GA accident causes, categorizing and explaining factors such as pilot skills, decision-making, training, and in-flight actions.", "Key Factual Information": "flight handling skills; poor judgement or airmanship; lack of training or experience; and omission of, or inappropriate, action", "Prefix reformulated": "What are the most common causes of GA accidents according to analysis?", "Prefix-intent": "analysis attributes the most common causes of these to", "Verification": "The KFI directly lists the common causes of accidents as identified by analysis, answering the query set up in the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 were 27 fatal accidents involving GA aircraft in 2007 , resulting in the loss of 48 lives . These compare with 16 accidents claiming a total of 19 lives the previous year , and although the 2007 statistics are higher than average , they are not exceptional . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 growth in Commercial Air Transport ( CAT ) has eroded the operational freedom of GA , both in the air and on the ground at larger airports . Difficulty with access to larger airports is compounded by a decline in the number of aerodromes generally , and existing sites are often threatened with closure and re @-@ development for more profitable uses . The UK planning system is designed to focus on local issues , and consideration of the national impact of GA operations is not within its remit . This makes aerodrome development difficult , often subjecting those that successfully negotiate the process to restrictions in use . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.", "role": "user" } ]

{ "Updated_Passage": null, "none": "", "results": null }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Airspace is shared by CAT , military and GA users . It is divided into controlled airspace , in which aircraft must always be under the control of an air traffic controller , and uncontrolled airspace , in which aircraft can operate autonomously . Although GA flights can under certain conditions enter controlled airspace , they operate mainly outside of it . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Controlled airspace is essential for the provision of a known air traffic environment necessary for the safe operation of CAT . A CAA review found that \" mixing [ commercial ] operations with other users is considered undesirable , even untenable \" by commercial operators . However this position has resulted in extensive Class A controlled airspace with complex boundaries , including some running down to the ground , prohibiting VFR access to airspace , resulting in high numbers of GA flights operating close to the borders of controlled airspace who could not get formal receipt of an air traffic service . Coupled with pilot navigation errors , hundreds of airspace infringements have been recorded every year . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Increases in the number of CAT operations , and in the number of airports they operate from , has resulted in a corresponding increase in Class A controlled airspace . Between 1997 and 2006 this area grew in size from 13 per cent of all airspace to 22 per cent nationally , and from 24 per cent to 43 per cent in airspace above England and Wales , leading to a perception within the GA community of being squeezed out . There are particular problems for GA around large airports , where Class A controlled airspace extends to ground level . The concentration of commercial operations and high demand for GA in the South East of England have also resulted in extensive areas of Class A controlled airspace there , which serve to channel uncontrolled GA operations through high @-@ collision @-@ risk hot spots . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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|>Increases in the number of CAT operations , and in the number of airports they operate from , has resulted in a corresponding increase in<|end-PI|> <|retrieval|> <|start-KFI|>Class A controlled airspace<|end-KFI|> . <|start-PI|>Between 1997 and 2006 this area grew in size from 13 per cent of all airspace to<|end-PI|> <|retrieval|> <|start-KFI|>22 per cent nationally , and from 24 per cent to 43 per cent in airspace above England and Wales<|end-KFI|> , leading to a perception within the GA community of being squeezed out . <|start-PI|>There are particular problems for GA around large airports , where<|end-PI|> <|retrieval|> <|start-KFI|>Class A controlled airspace extends to ground level<|end-KFI|> . <|start-PI|>The concentration of commercial operations and high demand for GA in the South East of England have also resulted in<|end-PI|> <|retrieval|> <|start-KFI|>extensive areas of Class A controlled airspace there , which serve to channel uncontrolled GA operations through high @-@ collision @-@ risk hot spots<|end-KFI|> .", "none": null, "results": [ { "Explanation": "The PI sets up the context for the increase in CAT operations and airports, while the KFI provides the specific result of this increase. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.", "Ideal Knowledge": "Detailed information about the relationship between Commercial Air Transport (CAT) operations, airport numbers, and the expansion of Class A controlled airspace, including historical data and trends.", "Key Factual Information": "Class A controlled airspace", "Prefix reformulated": "What has increased as a result of more CAT operations and airports?", "Prefix-intent": "Increases in the number of CAT operations , and in the number of airports they operate from , has resulted in a corresponding increase in", "Verification": "The KFI directly provides the information about what has increased, as prompted by the PI." }, { "Explanation": "The PI sets up the timeframe and initial percentage, while the KFI provides the specific growth figures. The <|retrieval|> token is inserted to indicate where additional statistical information might be needed.", "Ideal Knowledge": "Statistical data on the growth of Class A controlled airspace in different regions of the UK between 1997 and 2006, including national figures and specific data for England and Wales.", "Key Factual Information": "22 per cent nationally , and from 24 per cent to 43 per cent in airspace above England and Wales", "Prefix reformulated": "How much did Class A controlled airspace grow nationally between 1997 and 2006?", "Prefix-intent": "Between 1997 and 2006 this area grew in size from 13 per cent of all airspace to", "Verification": "The KFI directly provides the growth percentages for Class A controlled airspace, as prompted by the PI." }, { "Explanation": "The PI introduces the context of problems for GA near large airports, while the KFI specifies the particular issue. The <|retrieval|> token is inserted to indicate where additional information about airspace classification might be needed.", "Ideal Knowledge": "Information on the structure of airspace around large airports, focusing on the extent of Class A controlled airspace and its impact on General Aviation operations.", "Key Factual Information": "Class A controlled airspace extends to ground level", "Prefix reformulated": "What is the issue for General Aviation (GA) around large airports?", "Prefix-intent": "There are particular problems for GA around large airports , where", "Verification": "The KFI directly describes the specific problem for GA around large airports, as prompted by the PI." }, { "Explanation": "The PI sets up the context of concentrated operations and demand, while the KFI provides the specific result and its implications. The <|retrieval|> token is inserted to indicate where additional information about airspace management and safety might be needed.", "Ideal Knowledge": "Detailed analysis of airspace structure in South East England, including the extent of Class A controlled airspace, its impact on GA operations, and identification of high-risk areas for potential collisions.", "Key Factual Information": "extensive areas of Class A controlled airspace there , which serve to channel uncontrolled GA operations through high @-@ collision @-@ risk hot spots", "Prefix reformulated": "What has resulted from the concentration of commercial operations and high GA demand in South East England?", "Prefix-intent": "The concentration of commercial operations and high demand for GA in the South East of England have also resulted in", "Verification": "The KFI directly provides the result of the concentration of operations and demand, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Regional airports , such as Edinburgh Airport , have experienced strong growth in CAT operations in recent years . These operations are commercially and operationally incompatible with GA , and although there is no evidence of deliberate discrimination , the effect has been to discourage or exclude it . GA aircraft are being subject to significant increases in charges , including the imposition of handling fees in some cases . Some airports restrict or deny GA parking , and others limit or refuse certain GA activity . As a result , light GA aircraft are now rarely or never seen at large , busy international airports such as Heathrow , Stansted , Gatwick and Manchester . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 this de facto loss of facilities , the number of aerodromes in the UK has been in decline over the last 50 years , as a result of increasing urbanisation and the closure of airfields built during WWII . Alternative and more profitable uses for land can also lead to existing aerodromes being threatened with closure , for example North Weald , or actually being closed , as happened to Ipswich Airport. and Bristol Filton Airport . Referring to the importance of a \" functioning national network of GA airfields \" , especially where GA performs an air transport role , the CAA states that \" there could be cause for concern if a significant further loss of airfields were to continue , especially if crucial nodes on the transport network were to be lost . \" \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 this de facto loss of facilities , <|start-PI|> the number of aerodromes in the UK has been in decline over the last <|end-PI|> <|retrieval|> <|start-KFI|> 50 years , as a result of increasing urbanisation and the closure of airfields built during WWII <|end-KFI|> . <|start-PI|> Alternative and more profitable uses for land can also lead to existing aerodromes being <|end-PI|> <|retrieval|> <|start-KFI|> threatened with closure , for example North Weald , or actually being closed , as happened to Ipswich Airport. and Bristol Filton Airport <|end-KFI|> . <|start-PI|> Referring to the importance of a \" functioning national network of GA airfields \" , especially where GA performs an air transport role , the CAA states that <|end-PI|> <|retrieval|> <|start-KFI|> \" there could be cause for concern if a significant further loss of airfields were to continue , especially if crucial nodes on the transport network were to be lost . \" <|end-KFI|>", "none": null, "results": [ { "Explanation": "The PI sets up the context of declining aerodromes in the UK, while the KFI provides the specific timeframe and reasons. The <|retrieval|> token is inserted to indicate where additional historical information might be needed.", "Ideal Knowledge": "A comprehensive report on the historical trends of aerodromes in the UK, including statistics on their numbers over the past 50 years, reasons for closures, and the impact of urbanization and post-WWII developments.", "Key Factual Information": "50 years, as a result of increasing urbanisation and the closure of airfields built during WWII", "Prefix reformulated": "How long has the number of aerodromes in the UK been declining?", "Prefix-intent": "the number of aerodromes in the UK has been in decline over the last", "Verification": "The KFI directly provides the timeframe and reasons for the decline mentioned in the PI." }, { "Explanation": "The PI introduces the concept of alternative land uses affecting aerodromes, and the KFI provides specific examples. The <|retrieval|> token is inserted to indicate where additional information about affected aerodromes might be needed.", "Ideal Knowledge": "A detailed list of UK aerodromes that have been threatened with closure or actually closed due to land repurposing, including specific cases like North Weald, Ipswich Airport, and Bristol Filton Airport, with reasons for their closure or threats.", "Key Factual Information": "threatened with closure, for example North Weald, or actually being closed, as happened to Ipswich Airport. and Bristol Filton Airport", "Prefix reformulated": "What can happen to existing aerodromes due to alternative and more profitable land uses?", "Prefix-intent": "Alternative and more profitable uses for land can also lead to existing aerodromes being", "Verification": "The KFI directly provides examples of aerodromes affected by alternative land uses, as prompted by the PI." }, { "Explanation": "The PI sets up the context of the CAA's view on GA airfields, and the KFI provides the direct quote expressing their concern. The <|retrieval|> token is inserted to indicate where additional information about CAA policies or statements might be needed.", "Ideal Knowledge": "Official CAA reports or statements on the importance of general aviation (GA) airfields in the UK, including their role in the national transport network and concerns about potential closures.", "Key Factual Information": "\"there could be cause for concern if a significant further loss of airfields were to continue, especially if crucial nodes on the transport network were to be lost.\"", "Prefix reformulated": "What does the CAA state about the importance of a functioning national network of GA airfields?", "Prefix-intent": "Referring to the importance of a \"functioning national network of GA airfields\", especially where GA performs an air transport role, the CAA states that", "Verification": "The KFI directly quotes the CAA's statement regarding the importance of GA airfields, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 planning system is critical to the viability and operation of GA aerodromes . With many cities lacking scheduled air transport services between them , and with GA access to commercial airports becoming increasingly difficult and expensive , a viable network of aerodromes supporting GA air transport operations is regarded as an important national issue . However , there is no unified national planning policy specific to GA aerodromes , and planning decisions relating to these are based on local issues that are not required to consider the national impact . Because aircraft are excluded from noise control legislation , the only recourse for people affected by aircraft noise is through the planning process , and this issue is the principal factor on which the majority of planning decisions relating to GA land use are made . GA is a specialist subject often unfamiliar to Local Planning Authorities , and most planning decisions relating to GA either refuse permission , or grant it with restrictive conditions . Little Gransden is just one example of a GA airfield required to comply with planning restrictions on the number of movements permitted , thereby inhibiting further development . Such restrictions , if poorly conceived , can make GA operations unviable or even unsafe . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Public opinion towards aviation generally is worsening , based on increasing environmental concerns relating to emissions and noise , and private flying has been criticised by respondents to a government consultation on aircraft noise as a frivolous or selfish activity . In terms of environmental complaints and enquiries made to the CAA that relate specifically to GA , noise is \" by far \" the most common subject . Half of the 2 @,@ 000 noise complaints made annually to the CAA concern GA operations , most of which relate to aerobatics , helicopters using private sites , air balloon incidents , parachute dropping , and alleged low flying . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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|>Public opinion towards aviation generally is worsening , based on<|end-PI|> <|retrieval|> <|start-KFI|>increasing environmental concerns relating to emissions and noise<|end-KFI|> , and <|start-PI|>private flying has been criticised by respondents to a government consultation on aircraft noise as<|end-PI|> <|retrieval|> <|start-KFI|>a frivolous or selfish activity<|end-KFI|> . <|start-PI|>In terms of environmental complaints and enquiries made to the CAA that relate specifically to GA , noise is<|end-PI|> <|retrieval|> <|start-KFI|>\" by far \" the most common subject<|end-KFI|> . <|start-PI|>Half of the 2 @,@ 000 noise complaints made annually to the CAA concern<|end-PI|> <|retrieval|> <|start-KFI|>GA operations , most of which relate to aerobatics , helicopters using private sites , air balloon incidents , parachute dropping , and alleged low flying<|end-KFI|> .", "none": null, "results": [ { "Explanation": "The PI introduces the topic of worsening public opinion on aviation, while the KFI provides the specific reasons. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.", "Ideal Knowledge": "A comprehensive analysis of public sentiment towards aviation, including statistics and trends related to environmental concerns, specifically emissions and noise pollution.", "Key Factual Information": "increasing environmental concerns relating to emissions and noise", "Prefix reformulated": "What is causing public opinion towards aviation to worsen?", "Prefix-intent": "Public opinion towards aviation generally is worsening , based on", "Verification": "The KFI directly provides the reasons for worsening public opinion on aviation, as prompted by the PI." }, { "Explanation": "The PI sets up the context of criticism towards private flying, and the KFI provides the specific nature of that criticism. The <|retrieval|> token is inserted to indicate where additional information from the consultation might be needed.", "Ideal Knowledge": "Detailed results from government consultations on aircraft noise, focusing on public perceptions of private flying and its social impact.", "Key Factual Information": "a frivolous or selfish activity", "Prefix reformulated": "How do respondents to a government consultation on aircraft noise view private flying?", "Prefix-intent": "private flying has been criticised by respondents to a government consultation on aircraft noise as", "Verification": "The KFI directly provides the criticism of private flying as mentioned in the government consultation, as prompted by the PI." }, { "Explanation": "The PI introduces the topic of environmental complaints to the CAA about General Aviation, and the KFI emphasizes that noise is the primary concern. The <|retrieval|> token is inserted to indicate where specific data might be needed to support this claim.", "Ideal Knowledge": "Statistical breakdown of environmental complaints and enquiries made to the Civil Aviation Authority (CAA) regarding General Aviation, with a focus on noise-related issues.", "Key Factual Information": "\" by far \" the most common subject", "Prefix reformulated": "What is the most common subject of environmental complaints and enquiries made to the CAA regarding General Aviation?", "Prefix-intent": "In terms of environmental complaints and enquiries made to the CAA that relate specifically to GA , noise is", "Verification": "The KFI directly answers the question about the most common subject of complaints, as prompted by the PI." }, { "Explanation": "The PI sets up the context of noise complaints to the CAA, and the KFI provides specific information about the types of GA operations involved. The <|retrieval|> token is inserted to indicate where additional statistical information might be needed to support these claims.", "Ideal Knowledge": "Detailed breakdown of noise complaints received by the CAA, categorized by type of General Aviation operation, with specific focus on aerobatics, helicopters, air balloons, parachute dropping, and low flying incidents.", "Key Factual Information": "GA operations , most of which relate to aerobatics , helicopters using private sites , air balloon incidents , parachute dropping , and alleged low flying", "Prefix reformulated": "What type of aviation operations do half of the 2,000 annual noise complaints to the CAA concern?", "Prefix-intent": "Half of the 2 @,@ 000 noise complaints made annually to the CAA concern", "Verification": "The KFI directly provides information about the types of GA operations that are the subject of noise complaints, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Planning guidance on aircraft noise advises that \" in some circumstances the public perceive general aircraft noise levels as more disturbing than similar levels around major airports . \" This is a result of the tonal characteristics of light aircraft engines and the activities they are engaged in , including : repetitive circuit flying at low @-@ altitude near an aerodrome , during which aircraft are audible for long periods ; slow climbing aircraft engaged in parachute drop or glider tug activities concentrated around the drop zone or aerodrome , also audible for long periods ; erratic and repetitive engine noise from aircraft engaged in aerobatics ; and piston @-@ engines on full power in areas of low background noise , leading to the perception that such noise is more intrusive . In an attempt to alleviate these problems , the majority of aerodromes implement noise abatement procedures designed to route aircraft away from noise sensitive areas , and more than 50 are required by the government to provide consultative facilities in which local concerns can be raised with aerodrome operators . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.", "role": "user" } ]

{ "Updated_Passage": "Planning guidance on aircraft noise advises that <|start-PI|> Planning guidance on aircraft noise advises that <|end-PI|> <|retrieval|> <|start-KFI|> in some circumstances the public perceive general aircraft noise levels as more disturbing than similar levels around major airports <|end-KFI|> . <|start-PI|> This is a result of the tonal characteristics of light aircraft engines and the activities they are engaged in , including : <|end-PI|> <|retrieval|> <|start-KFI|> repetitive circuit flying at low @-@ altitude near an aerodrome , during which aircraft are audible for long periods ; slow climbing aircraft engaged in parachute drop or glider tug activities concentrated around the drop zone or aerodrome , also audible for long periods ; erratic and repetitive engine noise from aircraft engaged in aerobatics ; and piston @-@ engines on full power in areas of low background noise , leading to the perception that such noise is more intrusive <|end-KFI|> . <|start-PI|> In an attempt to alleviate these problems , the majority of aerodromes implement <|end-PI|> <|retrieval|> <|start-KFI|> noise abatement procedures designed to route aircraft away from noise sensitive areas <|end-KFI|> , and <|start-PI|> more than 50 are required by the government to provide <|end-PI|> <|retrieval|> <|start-KFI|> consultative facilities in which local concerns can be raised with aerodrome operators <|end-KFI|> .", "none": null, "results": [ { "Explanation": "The PI introduces the topic of aircraft noise guidance, while the KFI provides specific advice. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the statement.", "Ideal Knowledge": "Detailed information on aircraft noise regulations, public perception studies, and comparisons between general aviation and major airport noise levels.", "Key Factual Information": "in some circumstances the public perceive general aircraft noise levels as more disturbing than similar levels around major airports", "Prefix reformulated": "What does planning guidance on aircraft noise advise?", "Prefix-intent": "Planning guidance on aircraft noise advises that", "Verification": "The KFI directly provides the advice mentioned in the PI regarding public perception of aircraft noise." }, { "Explanation": "The PI sets up an explanation for the perceived disturbance, and the KFI provides a comprehensive list of contributing factors. The <|retrieval|> token is inserted to indicate where additional information about these activities might be needed.", "Ideal Knowledge": "Detailed explanations of various light aircraft activities, their noise profiles, and how they affect public perception, including technical information on engine types and flight patterns.", "Key Factual Information": "repetitive circuit flying at low @-@ altitude near an aerodrome , during which aircraft are audible for long periods ; slow climbing aircraft engaged in parachute drop or glider tug activities concentrated around the drop zone or aerodrome , also audible for long periods ; erratic and repetitive engine noise from aircraft engaged in aerobatics ; and piston @-@ engines on full power in areas of low background noise , leading to the perception that such noise is more intrusive", "Prefix reformulated": "What activities contribute to the perceived disturbance of light aircraft noise?", "Prefix-intent": "This is a result of the tonal characteristics of light aircraft engines and the activities they are engaged in , including :", "Verification": "The KFI directly lists the activities and characteristics that contribute to the perceived disturbance, as prompted by the PI." }, { "Explanation": "The PI introduces the topic of problem-solving measures, and the KFI provides specific information about noise abatement procedures. The <|retrieval|> token is inserted to indicate where additional information about these procedures might be needed.", "Ideal Knowledge": "Detailed information on various noise abatement procedures used by aerodromes, including specific routing strategies and their effectiveness in reducing noise impact on sensitive areas.", "Key Factual Information": "noise abatement procedures designed to route aircraft away from noise sensitive areas", "Prefix reformulated": "What do aerodromes implement to alleviate noise problems?", "Prefix-intent": "In an attempt to alleviate these problems , the majority of aerodromes implement", "Verification": "The KFI directly describes the measures implemented by aerodromes to address noise issues, as prompted by the PI." }, { "Explanation": "The PI sets up a query about government requirements, and the KFI provides specific information about consultative facilities. The <|retrieval|> token is inserted to indicate where additional information about these requirements might be needed.", "Ideal Knowledge": "Information on government regulations for aerodromes, specifically detailing requirements for public consultation and community engagement regarding noise concerns.", "Key Factual Information": "consultative facilities in which local concerns can be raised with aerodrome operators", "Prefix reformulated": "What are more than 50 aerodromes required by the government to provide?", "Prefix-intent": "more than 50 are required by the government to provide", "Verification": "The KFI directly states what the government requires from aerodromes, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n SMS Zrínyi ( \" His Majesty 's ship Zrínyi \" ) was a Radetzky @-@ class pre @-@ dreadnought battleship ( Schlachtschiff ) of the Austro @-@ Hungarian Navy ( K.u.K. Kriegsmarine ) , named for the Zrinski , a noble Croatian family . Zrínyi and her sisters , Erzherzog Franz Ferdinand and Radetzky , were the last pre @-@ dreadnoughts built by the Austro @-@ Hungarian Navy . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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|>SMS Zrínyi ( \" His Majesty 's ship Zrínyi \" ) was<|end-PI|> <|retrieval|> <|start-KFI|>a Radetzky @-@ class pre @-@ dreadnought battleship ( Schlachtschiff ) of the Austro @-@ Hungarian Navy ( K.u.K. Kriegsmarine )<|end-KFI|> , <|start-PI|>named for<|end-PI|> <|retrieval|> <|start-KFI|>the Zrinski , a noble Croatian family<|end-KFI|> . <|start-PI|>Zrínyi and her sisters , Erzherzog Franz Ferdinand and Radetzky , were<|end-PI|> <|retrieval|> <|start-KFI|>the last pre @-@ dreadnoughts built by the Austro @-@ Hungarian Navy<|end-KFI|> .", "none": null, "results": [ { "Explanation": "The PI introduces SMS Zrínyi and sets up a query about its classification, while the KFI provides specific details about its type and affiliation. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.", "Ideal Knowledge": "Detailed information about naval ship classifications, specifically focusing on pre-dreadnought battleships and the Radetzky-class ships of the Austro-Hungarian Navy.", "Key Factual Information": "a Radetzky @-@ class pre @-@ dreadnought battleship ( Schlachtschiff ) of the Austro @-@ Hungarian Navy ( K.u.K. Kriegsmarine )", "Prefix reformulated": "What type of ship was SMS Zrínyi?", "Prefix-intent": "SMS Zrínyi ( \" His Majesty 's ship Zrínyi \" ) was", "Verification": "The KFI directly provides information about the type and class of the ship SMS Zrínyi, as prompted by the PI." }, { "Explanation": "The PI sets up a query about the ship's name origin, and the KFI provides the specific family name and their status. The <|retrieval|> token is inserted to indicate where additional information about the Zrinski family might be needed.", "Ideal Knowledge": "Historical information about the Zrinski family, their significance in Croatian history, and their connection to naval naming conventions in the Austro-Hungarian Empire.", "Key Factual Information": "the Zrinski , a noble Croatian family", "Prefix reformulated": "Who was SMS Zrínyi named after?", "Prefix-intent": "SMS Zrínyi ( \" His Majesty 's ship Zrínyi \" ) was a Radetzky @-@ class pre @-@ dreadnought battleship ( Schlachtschiff ) of the Austro @-@ Hungarian Navy ( K.u.K. Kriegsmarine ) , named for", "Verification": "The KFI directly provides information about the namesake of SMS Zrínyi, as prompted by the PI." }, { "Explanation": "The PI introduces Zrínyi and its sister ships, setting up a query about their importance, while the KFI provides the specific historical context. The <|retrieval|> token is inserted to indicate where additional information about naval history might be needed.", "Ideal Knowledge": "Detailed information about the naval history of the Austro-Hungarian Empire, focusing on the transition from pre-dreadnought to dreadnought battleships, and the significance of the Radetzky-class ships in this context.", "Key Factual Information": "the last pre @-@ dreadnoughts built by the Austro @-@ Hungarian Navy", "Prefix reformulated": "What was the significance of Zrínyi, Erzherzog Franz Ferdinand, and Radetzky in the Austro-Hungarian Navy?", "Prefix-intent": "Zrínyi and her sisters , Erzherzog Franz Ferdinand and Radetzky , were", "Verification": "The KFI directly provides information about the historical significance of these ships in the Austro-Hungarian Navy, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n During World War I , Zrínyi saw action in the Adriatic Sea . She served with the Second Division of the Austro @-@ Hungarian Navy 's battleships and shelled Senigallia as part of the bombardment of the key seaport of Ancona , Italy , during May 1915 . However , Allied control of the Strait of Otranto meant that the Austro @-@ Hungarian Navy was , for all intents and purposes , effectively bottled up in the Adriatic . Nonetheless , the presence of the Zrínyi and other battleships tied down a substantial force of Allied ships . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n With the war going against the Austrians by the end of 1918 , Zrínyi was prepared to be transferred to the new State of Slovenes , Croats and Serbs . On 10 November 1918 — just one day before the end of the war , navy officers sailed the battleship out of Pola ( Pula ) and eventually surrendered to a squadron of American submarine chasers . Following the handover to the United States Navy , she was briefly designated USS Zrínyi . In the Treaty of Saint @-@ Germain @-@ en @-@ Laye , the transfer was not recognized ; instead , Zrínyi was given to Italy and broken up for scrap . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.", "role": "user" } ]

{ "Updated_Passage": null, "none": "", "results": null }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Zrínyi was built at the Stabilimento Tecnico Triestino dockyard in Trieste , the same place where her sister ships were built earlier . She was laid down on 15 November 1908 and launched from the slipway on 12 April 1910 . The teak used on Zrínyi 's deck was the only material Austria @-@ Hungary had to purchase abroad to build the ship . The ship was completed by 15 July 1911 , and on 22 November 1911 she was commissioned into the fleet . She was the last ship of the class to be completed and had a crew of 880 to 890 officers and men . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Zrínyi was 138 @.@ 8 m ( 455 ft 4 in ) long , and had a beam of 24 @.@ 6 m ( 80 ft 8 in ) and a draft of 8 @.@ 1 m ( 26 ft 9 in ) . She displaced 14 @,@ 508 long tons ( 14 @,@ 741 t ) normally , and up to 15 @,@ 845 long tons ( 16 @,@ 099 t ) with a full combat load . She was powered by two @-@ shaft four @-@ cylinder vertical triple expansion engines rated at 19 @,@ 800 indicated horsepower . The ship had a top speed of 20 @.@ 5 knots ( 38 @.@ 0 km / h ; 23 @.@ 6 mph ) . Zrínyi was the first warship in the Austro @-@ Hungarian Navy to use fuel oil to supplement her 12 Yarrow @-@ type coal @-@ fired boilers . She had a maximum range of 4 @,@ 000 nautical miles ( 7 @,@ 400 km ; 4 @,@ 600 mi ) at a cruising speed of 10 knots ( 19 km / h ; 12 mph ) . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 ship 's primary armament consisted of four 30 @.@ 5 cm ( 12 in ) 45 @-@ caliber guns in two twin gun turrets . This was augmented by a heavy secondary battery of eight 24 cm ( 9 @.@ 4 in ) guns in four wing turrets . The tertiary battery consisted of twenty 10 cm L / 50 guns in casemated single mounts , four 47 mm ( 1 @.@ 85 in ) L / 44 and one 47 mm L / 33 quick @-@ firing guns . Furthermore , the ship 's boats were equipped with two 66 mm ( 2 @.@ 6 in ) landing guns for operations shore . Three 45 cm ( 17 @.@ 7 in ) torpedo tubes were also carried , one on each broadside and one in the stern . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 ship was assigned to the Austro @-@ Hungarian Fleet 's 1st Battle Squadron after her 1911 commissioning . In 1912 , Zrínyi and her two sister ships conducted two training cruises into the eastern Mediterranean Sea . On the second cruise into the Aegean Sea , conducted from November to December , Zrínyi and her sister ships were accompanied by the cruiser SMS Admiral Spaun and a pair of destroyers . After returning to Pola , the entire fleet mobilized for possible hostilities , as tensions flared in the Balkans . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 1913 , Zrínyi participated in an international naval demonstration in the Ionian Sea to protest the Balkan Wars . Ships from other navies included in the demonstration were the British pre @-@ dreadnought HMS King Edward VII , the Italian pre @-@ dreadnought Ammiraglio di Saint Bon , the French armored cruiser Edgar Quinet , and the German light cruiser SMS Breslau . The most important action of the combined flotilla , which was under the command of British Admiral Cecil Burney , was to blockade the Montenegrin coast . The goal of the blockade was to prevent Serbian reinforcements from supporting the siege at Scutari , where Montenegro had besieged a combined force of Albanians and Ottomans . Pressured by the international blockade , Serbia withdrew its army from Scutari , which was subsequently occupied by a joint Allied ground force . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n During that year , the first of four new dreadnoughts , SMS Viribus Unitis , that made up the Tegetthoff class — the only dreadnoughts built for the Austro @-@ Hungarian Navy — came into active service . With the commissioning of these dreadnoughts , Zrínyi and her sisters were moved from the 1st Division to the 2nd Division of the 1st Battle Squadron . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 that time of the assassination of Archduke Franz Ferdinand of Austria on 28 June 1914 , the battleships in the Austro @-@ Hungarian Navy consisted of the Radetzky class , the Tegetthoff class ( which still had one ship , SMS Szent István , under construction ) , the Erzherzog Karl class and finally , the older Habsburg class . Along with the remainder of the Austro @-@ Hungarian Navy , Zrínyi was mobilized in late July 1914 to support the flight of SMS Goeben and SMS Breslau . The two German ships broke out of Messina , which was surrounded by the British navy and reached Turkey . The flotilla had advanced as far south as Brindisi in southeastern Italy when news of the successful breakout reached Vienna . The Austro @-@ Hungarian ships were then recalled before seeing action . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 23 May 1915 , between two and four hours after news of the Italian declaration of war reached the main Austro @-@ Hungarian naval base at Pola , Zrínyi and the rest of the fleet departed to bombard the Italian and Montenegrin coast . Their focus was on the important naval base at Ancona , and later the coast of Montenegro . The bombardment of Montenegro was part of the larger Austro @-@ Hungarian campaign against the Kingdoms of Montenegro and Serbia , who were members of the Entente , during the first half of 1915 . The attack on Ancona was an immense success , and the ships were unopposed during the operation . The bombardment of the province and the surrounding area resulted in the destruction of an Italian steamer in the port of Ancona itself , and an Italian destroyer , Turbine , was severely damaged further south . On the shore , the infrastructure of the port of Ancona , as well as the surrounding towns , were severely damaged . The railroad yard in Ancona , as well as the port facilities in the town , were damaged or destroyed . The local shore batteries were also suppressed . During the bombardment , Zrínyi also helped to destroy a train , a railway station , and a bridge at Senigallia . Additional targets that were damaged or destroyed included wharves , warehouses , oil tanks , radio stations , and the local barracks . Sixty @-@ three Italians , both civilians and military personnel , were killed in the bombardment . By the time Italian ships from Taranto and Brindisi arrived on the scene , the Austro @-@ Hungarians were safely back in Pola . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 objective of the bombardment of Ancona was to delay the Italian Army from deploying its forces along the border with Austria @-@ Hungary by destroying critical transportation systems . The surprise attack on Ancona succeeded in delaying the Italian deployment to the Alps for two weeks . This delay gave Austria @-@ Hungary valuable time to strengthen its Italian border and re @-@ deploy some of its troops from the Eastern and Balkan fronts . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 the attack on Ancona , the Austro @-@ Hungarian battleships were largely confined to Pola for the duration of the war . Their operations were limited by Admiral Anton Haus , the commander of the Austro @-@ Hungarian Navy , who believed that he would need to husband his ships to counter any Italian attempt to seize the Dalmatian coast . Since coal was diverted to the newer Tegetthoff @-@ class battleships , the remainder of the war saw Zrínyi and the rest of the Austro @-@ Hungarian Navy acting as a fleet in being . This resulted in the Allied blockade of the Otranto Strait . With his fleet blockaded in the Adriatic Sea , and with a shortage of coal , Haus followed a strategy based on mines and submarines designed to reduce the numerical superiority of the Allied navies . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n After the Austro @-@ Hungarian Empire collapsed in 1918 , the Austrians wanted to turn the fleet over to the newly created State of Slovenes , Croats and Serbs ( later to become a part of the Kingdom of Yugoslavia ) in order to prevent the Italians from claiming the ships as spoils of war . However , the victorious Allies refused to acknowledge the conversations between the Austrians and the south Slavs and , in due course , reallocated the ships . The ship had been boarded by a scratch Yugoslav crew on 10 November 1918 , one day before the Armistice , and had left Pola with her sister ship , Radetzky . They were soon spotted by heavy Italian ships , so the two battleships hoisted American flags and sailed south along the Adriatic coast to Castelli Bay near Spalato ( also known as Split ) . They appealed for American naval forces to meet them and accept their surrender , which a squadron of United States Navy ( USN ) submarine chasers in the area did . She had apparently been turned over to the fledgling south Slav state , as it was a Croat naval officer , Korvettenkapitän Marijan Polić , who presented the ship as a prize of war to representatives of the United States Navy on the afternoon of 22 November 1919 at Spalato ( Split ) in Dalmatia . Simultaneously she was commissioned as USS Zrínyi and Lieutenant E.E. Hazlett , USN , assumed command . The initial American complement consisted of four officers and 174 enlisted men — the latter entirely composed of United States Navy Reserve Force personnel . The ship remained at anchor at Spalato for nearly a year while the negotiations that would determine her ultimate fate dragged on . Only once did she apparently turn her engines over , and that occurred during a severe gale that struck Spalato on 9 February 1920 . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 the morning of 7 November 1920 , Zrínyi was decommissioned . USS Chattanooga took her in tow and , assisted by Brooks and Hovey , towed the battleship to Italy . Under the terms of the treaties of Versailles and St. Germain , Zrínyi was ultimately turned over to the Italian government at Venice . She was broken up for scrap later that year and into 1921 . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Geopyxis carbonaria is a species of fungus in the genus Geopyxis , family Pyronemataceae . First described to science in 1805 , and given its current name in 1889 , the species is commonly known as the charcoal loving elf @-@ cup , dwarf acorn cup , stalked bonfire cup , or pixie cup . The small , goblet @-@ shaped fruitbodies of the fungus are reddish @-@ brown with a whitish fringe and measure up to 2 cm ( 0 @.@ 8 in ) across . They have a short , tapered stalk . Fruitbodies are commonly found on soil where brush has recently been burned , sometimes in great numbers . The fungus is distributed throughout many temperate regions of the Northern Hemisphere . It is found in Europe , Turkey , and North America . Although it is primarily a saprotrophic species , feeding on the decomposing organic matter remaining after a fire , it also forms biotrophic associations with the roots of Norway spruce . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 fungus was first described scientifically in 1805 by Johannes Baptista von Albertini and Lewis David de Schweinitz as Peziza carbonaria . Mordecai Cubitt Cooke illustrated the fruitbodies , spores , and asci in his 1879 work Mycographia , seu Icones fungorum . Figures of fungi from all parts of the world . In 1889 , Pier Andrea Saccardo transferred the fungus to the genus Geopyxis , giving the species its current name . Pustularia carbonaria , published by Heinrich Rehm in 1884 , is a synonym of G. carbonaria . Louis @-@ Joseph Grélet proposed the variety Geopyxis carbonaria var. sessilis in 1937 , referring to forms producing fruitbodies without a stalk , but the taxon is not considered to have independent taxonomic significance . In 1860 Miles Berkeley and Moses Ashley Curtis described the species Peziza lepida from collections made in Japan as part of the North Pacific Exploring and Surveying Expedition ( 1853 – 1856 ) . This taxon was synonymized with G. carbonaria by Mien Rifai in 1968 , a taxonomic opinion corroborated by Donald Pfister about a decade 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n The specific epithet carbonaria derives from the Latin word for \" charcoal \" . Common names given to the fungus include \" charcoal loving elf @-@ cup \" , \" dwarf acorn cup \" , \" pixie cup \" , and the British Mycological Society approved \" stalked bonfire cup \" . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.", "role": "user" } ]

{ "Updated_Passage": null, "none": "", "results": null }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 fruitbodies ( ascocarps ) of Geopyxis carbonaris are cup shaped , 1 – 2 cm wide , and have fringed whitish margins . The inner spore @-@ bearing surface of the cup , the hymenium , is brick red and smooth , while the exterior surface is a dull yellow , and may be either smooth or have blister @-@ like spots ( pustules ) . The stipe is small ( 1 – 1 @.@ 5 mm long and 1 – 2 mm wide ) , whitish in color , and expands abruptly into the cup . The brownish flesh of the fungus is thin and brittle . It does not have any distinctive taste , but has an unpleasant smell when crushed in water . The edibility of the fungus is not known , but the fruitbodies are insubstantial and unlikely to be harvested for eating . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 mass , the spores are whitish . The spores are elliptical , smooth , hyaline , devoid of oil droplets ( eguttulate ) , and have dimensions of 13 – 18 by 7 – 9 µm . They are thin walled and germinate and grow rapidly in vitro in the absence of external stimuli . The asci are 190 – 225 by 9 – 10 µm . The paraphyses are slightly club @-@ shaped , unbranched , and have irregular orange @-@ brown granules , with tips up to 5 µm wide , and are not forked or lobed . The hypothecium , the layer of cells below the hymenium , is made of densely packed , small irregular cells . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 closely related vulcan elf cup ( Geopyxis vulcanalis ) has a pale orange to yellowish fruitbody that is deeply cup shaped before flattening in maturity , and its crushed flesh often has an odor of sulfur . It may be distinguished microscopically by its paraphyses , which lack the orange @-@ brown granules characteristic of G. carbonaria . It also has larger spores , measuring 14 – 22 by 8 – 11 µm . Unlike G. carbonaria , it grows on substrates other than burned wood , including mosses , and needle duff . Tarzetta cupularis , which grows habitats similar to G. carbonaria , is distinguished microscopically by its spores that contain two oil droplets . Other genera with similar species with which G. carbonaria may be confused in the field include Aleuria , Caloscypha , Melastiza , and Sowerbyella . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Geopyxis carbonaria is widespread on burned soil or charcoal in the spring and throughout the growing season . It is one of the most common pioneer species found on burned ground . The charred litter on the forest floor increases the underlying soil pH as well as the availability of minerals . Fruitbodies are produced from 16 to 139 weeks after a forest fire in areas with coniferous trees . Most fruitbodies are produced in the first year after a burn . The fungus prefers fruiting in microhabitats with thin postfire duff near standing burned tree trunks . Geopyxis carbonaria fruitbodies are often found in the same post @-@ fire stands as morels , although the former is usually more abundant . Because the pixie cup fruits earlier than morels , it may serve as an indicator of imminent morel fruiting . Other cup fungi often found fruiting in the same area as G. carbonaria include those from the genera Aleuria , Anthracobia , Peziza , and Tarzetta . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 fungus is found in Europe ( from where it was originally described ) , and is widespread throughout North America . The North American distribution extends north to Alaska . In 2010 , it was reported for the first time from Turkey . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Although primarily a saprotrophic fungus involved in the post @-@ fire breakdown of duff and coniferous roots , Geopyxis carbonaria has been shown to be capable of forming ectomycorrhizae with Norway spruce ( Picea abies ) . It had been demonstrated earlier in laboratory experiments that the fungus has a biotrophic interaction with lodgepole pine ( Pinus contorta ) . The hyphae of G. carbonaria were able to infect the cortex of the tree seedling , but did not penetrate the endodermis . These traits suggest that the fungus is a moderate pathogen , with limited ability to cause reductions in seed germination . Additionally , the fungus produces the enzyme polyphenol oxidase , and can break down the complex organic polymer lignin — features characteristic of saprotrophic fungi . The formation of a rudimentary Hartig net , a characteristic of mycorrhizal fungi , indicated that G. carbonaria might be capable of forming mutualistic relationships under the right conditions . Vrålstad and colleagues suggest that its below @-@ ground association with spruce roots protects it from physical damage in the event of a fire , and the extensive fruitbody production after a fire may reflect \" a successful fungal escape from a dying host where the fungus no longer can maintain its biotrophic association \" . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Large fruitings of the fungus are often associated with damage to the host tree , such as that which occurs with burning . A field study conducted in Norway demonstrated that fruit bodies were more likely to be found in areas that were heavily burned , compared to locations with light to moderate burning where the trees remained viable , or in clearcut areas . Fruiting was much denser in spruce forests — with up to 700 – 1000 fruitbodies per square meter — than in pine forests , where fruitbodies were sporadic . Fruitbodies grew by the millions in the year following the Yellowstone fires of 1988 . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 gold dollar or gold one @-@ dollar piece was a coin struck as a regular issue by the United States Bureau of the Mint from 1849 to 1889 . The coin had three types over its lifetime , all designed by Mint Chief Engraver James B. Longacre . The Type 1 issue had the smallest diameter of any United States coin ever minted . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 gold dollar or gold one @-@ dollar piece was a coin struck as a regular issue by the United States Bureau of the Mint from<|end-PI|> <|retrieval|> <|start-KFI|>1849 to 1889<|end-KFI|> . <|start-PI|>The coin had three types over its lifetime , all designed by<|end-PI|> <|retrieval|> <|start-KFI|>Mint Chief Engraver James B. Longacre<|end-KFI|> . <|start-PI|>The Type 1 issue had the smallest diameter of<|end-PI|> <|retrieval|> <|start-KFI|>any United States coin ever minted<|end-KFI|> .", "none": null, "results": [ { "Explanation": "The PI introduces the gold dollar and sets up a query about its production period, while the KFI provides the specific years. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.", "Ideal Knowledge": "A detailed timeline of U.S. coin production, specifically highlighting the period when the gold dollar was minted as a regular issue.", "Key Factual Information": "1849 to 1889", "Prefix reformulated": "When was the gold dollar or gold one-dollar piece struck as a regular issue by the United States Bureau of the Mint?", "Prefix-intent": "The gold dollar or gold one @-@ dollar piece was a coin struck as a regular issue by the United States Bureau of the Mint from", "Verification": "The KFI directly provides the time period requested in the PI." }, { "Explanation": "The PI mentions the three types of the coin and sets up a query about their designer, while the KFI provides the specific name and position. The <|retrieval|> token is inserted to indicate where additional information about the designer might be needed.", "Ideal Knowledge": "Biographical information about James B. Longacre, including his role as Mint Chief Engraver and his contributions to U.S. coin designs, particularly the gold dollar types.", "Key Factual Information": "Mint Chief Engraver James B. Longacre", "Prefix reformulated": "Who designed the three types of gold dollar coins?", "Prefix-intent": "The coin had three types over its lifetime , all designed by", "Verification": "The KFI directly provides the designer's name and position as requested in the PI." }, { "Explanation": "The PI introduces a unique feature of the Type 1 issue, and the KFI provides the specific information about its diameter being the smallest. The <|retrieval|> token is inserted to indicate where additional information about coin sizes might be needed.", "Ideal Knowledge": "A comprehensive list of U.S. coin diameters throughout history, highlighting the Type 1 gold dollar as the smallest, along with comparisons to other small coins.", "Key Factual Information": "any United States coin ever minted", "Prefix reformulated": "What was unique about the diameter of the Type 1 gold dollar issue?", "Prefix-intent": "The Type 1 issue had the smallest diameter of", "Verification": "The KFI directly provides information about the unique characteristic of the Type 1 issue's diameter as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 gold dollar had been proposed several times in the 1830s and 1840s , but was not initially adopted . Congress was finally galvanized into action by the increased supply of bullion caused by the California gold rush , and in 1849 authorized a gold dollar . In its early years , silver coins were being hoarded or exported , and the gold dollar found a ready place in commerce . Silver again circulated after Congress in 1853 required that new coins of that metal be made lighter , and the gold dollar became a rarity in commerce even before federal coins vanished from circulation because of the economic disruption caused by the American Civil War . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.", "role": "user" } ]

{ "Updated_Passage": null, "none": "", "results": null }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Gold did not again circulate in most of the nation until 1879 ; once it did , the gold dollar did not regain its place . In its final years , it was struck in small numbers , causing speculation by hoarders . It was also in demand to be mounted in jewelry . The regular issue gold dollar was last struck in 1889 ; the following year , Congress ended the series . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 proposing his plan for a mint and a coinage system , Secretary of the Treasury Alexander Hamilton in 1791 proposed that the one @-@ dollar denomination be struck both as a gold coin , and as one of silver , representative of the two metals which he proposed be made legal tender . Congress followed Hamilton 's recommendation only in part , authorizing a silver dollar , but no coin of that denomination in 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 1831 , the first gold dollar was minted , at the private mint of Christopher Bechtler in North Carolina . Much of the gold then being produced in the United States came from the mountains of North Carolina and Georgia , and the dollars and other small gold coins issued by Bechtler circulated through that region , and were now and then seen further away . Additional one @-@ dollar pieces were struck by August Bechtler , Christopher 's son . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Soon after the Bechtlers began to strike their private issues , Secretary of the Treasury Levi Woodbury became an advocate of having the Mint of the United States ( \" Mint \" , when described as an institution ) strike the one @-@ dollar denomination in gold . He was opposed by the Mint Director , Robert M. Patterson . Woodbury persuaded President Andrew Jackson to have pattern coins struck . In response , Patterson had Mint Second Engraver Christian Gobrecht break off work on the new design for the silver one @-@ dollar coin and work on a pattern for the gold dollar . Gobrecht 's design featured a Liberty cap surrounded by rays on one side , and a palm branch arranged in a circle with the denomination , date , and name of the country on the other . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Consideration was given to including the gold dollar as an authorized denomination in the revisionary legislation that became the Mint Act of 1837 . The Philadelphia newspaper Public Ledger , in December 1836 , supported a gold dollar , stating that \" the dollar is the smallest gold coin that would be convenient , and as it would be eminently so , neither silver nor paper should be allowed to take its place . \" Nevertheless , after Mint Director Patterson appeared before a congressional committee , the provision authorizing the gold dollar was deleted from the bill . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 January 1844 , North Carolina Representative James Iver McKay , the chairman of the Committee on Ways and Means , solicited the views of Director Patterson on the gold dollar . Patterson had more of Gobrecht 's pattern dollar struck to show to committee members , again advising against a coin that if issued would be only about a half inch ( 13 mm ) in diameter . He told Treasury Secretary John C. Spencer that the only gold coins of that size in commerce , the Spanish and Colombian half @-@ escudos , were unpopular and had not been struck for more than twenty years . This seemed to satisfy the committee as nothing more was done for the time , and when a gold dollar was proposed again in 1846 , McKay 's committee recommended against it . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Even before 1848 , record amounts of gold were flowing to American mints to be struck into coin , but the California Gold Rush vastly increased these quantities . This renewed calls for a gold dollar , as well as for a higher denomination than the eagle ( $ 10 piece ) , then the largest gold coin . In January 1849 , McKay introduced a bill for a gold dollar , which was referred to his committee . There was much discussion in the press about the proposed coin ; one newspaper published a proposal for an annular gold dollar , that is , with a hole in the middle to increase its small diameter . McKay amended his legislation to provide for a double eagle ( $ 20 gold coin ) and wrote to Patterson , who replied stating that the annular gold dollar would not work , and neither would another proposal to have dollar piece consisting of a gold plug in a silver coin . Nevertheless , Gobrecht 's successor as chief engraver , James B. Longacre , prepared patterns , including some with a square hole in the middle . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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|>Even before 1848, record amounts of gold were flowing to American mints to be struck into<|end-PI|> <|retrieval|> <|start-KFI|>coin, but the California Gold Rush vastly increased these quantities<|end-KFI|>. <|start-PI|>This renewed calls for a gold dollar, as well as for a higher denomination than the eagle ( $ 10 piece ), then<|end-PI|> <|retrieval|> <|start-KFI|>the largest gold coin<|end-KFI|>. <|start-PI|>In January 1849, McKay introduced a bill for a gold dollar, which<|end-PI|> <|retrieval|> <|start-KFI|>was referred to his committee<|end-KFI|>. There was much discussion in the press about the proposed coin; one newspaper published a proposal for an annular gold dollar, that is, with a hole in the middle to increase its small diameter. <|start-PI|>McKay amended his legislation to provide for a double eagle ( $ 20 gold coin ) and wrote to Patterson, who replied stating that<|end-PI|> <|retrieval|> <|start-KFI|>the annular gold dollar would not work, and neither would another proposal to have dollar piece consisting of a gold plug in a silver coin<|end-KFI|>. <|start-PI|>Nevertheless, Gobrecht's successor as chief engraver, James B. Longacre,<|end-PI|> <|retrieval|> <|start-KFI|>prepared patterns, including some with a square hole in the middle<|end-KFI|>.", "none": null, "results": [ { "Explanation": "The PI sets up the context of gold flowing to American mints, while the KFI provides the specific information about coin production and the impact of the California Gold Rush. The <|retrieval|> token is inserted to indicate where additional information about pre-1848 gold coin production might be needed.", "Ideal Knowledge": "Information about gold coin production in American mints before 1848, including statistics on gold quantities and the impact of the California Gold Rush on coin production.", "Key Factual Information": "coin, but the California Gold Rush vastly increased these quantities", "Prefix reformulated": "What were record amounts of gold being struck into at American mints before 1848?", "Prefix-intent": "Even before 1848, record amounts of gold were flowing to American mints to be struck into", "Verification": "The KFI directly relates to the PI by confirming that gold was being struck into coin and providing additional context about the California Gold Rush's impact." }, { "Explanation": "The PI introduces the context of calls for new gold coin denominations, while the KFI provides the specific information about the eagle being the largest gold coin. The <|retrieval|> token is inserted to indicate where additional information about gold coin denominations might be needed.", "Ideal Knowledge": "A list of gold coin denominations in use in the United States before 1849, with specific information about the eagle ($10 piece) being the largest denomination at that time.", "Key Factual Information": "the largest gold coin", "Prefix reformulated": "What was the largest gold coin denomination before the renewed calls for a gold dollar?", "Prefix-intent": "This renewed calls for a gold dollar, as well as for a higher denomination than the eagle ( $ 10 piece ), then", "Verification": "The KFI directly answers the question implied by the PI, confirming that the eagle was the largest gold coin at the time." }, { "Explanation": "The PI sets up the context of McKay introducing a bill for a gold dollar, while the KFI provides information about the bill's fate. The <|retrieval|> token is inserted to indicate where additional information about the legislative process might be needed.", "Ideal Knowledge": "Details about the legislative process for introducing new coin denominations in 1849, including information about McKay's role and the committee structure for handling such bills.", "Key Factual Information": "was referred to his committee", "Prefix reformulated": "What happened to the bill for a gold dollar introduced by McKay in January 1849?", "Prefix-intent": "In January 1849, McKay introduced a bill for a gold dollar, which", "Verification": "The KFI directly provides information about what happened to the bill mentioned in the PI." }, { "Explanation": "The PI sets up the context of McKay amending his legislation and writing to Patterson, while the KFI provides Patterson's specific response. The <|retrieval|> token is inserted to indicate where additional information about the proposed coin designs might be needed.", "Ideal Knowledge": "Detailed information about the correspondence between McKay and Patterson regarding proposed gold coin designs, including technical reasons why certain designs were deemed impractical.", "Key Factual Information": "the annular gold dollar would not work, and neither would another proposal to have dollar piece consisting of a gold plug in a silver coin", "Prefix reformulated": "What did Patterson reply about McKay's amended legislation for a double eagle?", "Prefix-intent": "McKay amended his legislation to provide for a double eagle ( $ 20 gold coin ) and wrote to Patterson , who replied stating that", "Verification": "The KFI directly provides Patterson's response to McKay's amended legislation, as prompted by the PI." }, { "Explanation": "The PI introduces James B. Longacre and his position, while the KFI provides specific information about his work on coin patterns. The <|retrieval|> token is inserted to indicate where additional information about Longacre's designs might be needed.", "Ideal Knowledge": "Information about James B. Longacre's role as chief engraver, including his work on various coin designs and patterns, particularly those with unique features like holes in the middle.", "Key Factual Information": "prepared patterns, including some with a square hole in the middle", "Prefix reformulated": "What did James B. Longacre, Gobrecht's successor as chief engraver, do?", "Prefix-intent": "Nevertheless, Gobrecht's successor as chief engraver, James B. Longacre,", "Verification": "The KFI directly provides information about Longacre's actions, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n McKay got his fellow Democrat , New Hampshire Senator Charles Atherton , to introduce the bill to authorize the gold dollar and the double eagle in the Senate on February 1 , 1849 — Atherton was chairman of the Senate Finance Committee . McKay introduced a version into the House on February 20 ; debate began the same day . The dollar was attacked by congressmen from the Whig Party , then in the minority , on the grounds that it would be too small , would be counterfeited and in bad light might be mistakenly spent as a half dime , the coins being similar in size . McKay did not respond substantively , but stated that if no one wanted these denominations , they would not be called for at the Mint , and would not be coined . Pennsylvania Representative Joseph Ingersoll , a Whig , spoke against the bill , noting that Patterson opposed the new denominations , and that the idea had been repeatedly turned down , whenever considered . Another Whig , Massachusetts 's Charles Hudson , related that Patterson had sent a real and a counterfeit gold dollar to his committee and the majority of members had been unable to tell the difference . McKay made no answer to these claims , but others did , including New York Congressman Henry Nicoll , who assured the House that the counterfeiting allegations were greatly exaggerated . The point was , he indicated , that the double eagle and gold dollar were wanted by the public , and , in the case of the gold dollar could help money circulate in small communities where banknotes were not accepted . Connecticut Representative John A. Rockwell , a Whig , tried to table the bill , but his motion was defeated . The bill passed easily , and met only minimal opposition in the Senate , becoming law on March 3 , 1849 . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 officers at the Philadelphia Mint , including Chief Coiner Franklin Peale , were mostly the friends and relations of Director Patterson . The outsider in their midst was Chief Engraver James B. Longacre , successor to Gobrecht ( who had died in 1844 ) . A former copper @-@ plate engraver , Longacre had been appointed through the political influence of South Carolina Senator John C. Calhoun . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 officers at the Philadelphia Mint, including <|start-PI|>Chief Coiner<|end-PI|> <|retrieval|> <|start-KFI|>Franklin Peale<|end-KFI|>, <|start-PI|>were mostly<|end-PI|> <|retrieval|> <|start-KFI|>the friends and relations of Director Patterson<|end-KFI|>. <|start-PI|>The outsider in their midst was Chief Engraver<|end-PI|> <|retrieval|> <|start-KFI|>James B. Longacre, successor to Gobrecht (who had died in 1844)<|end-KFI|>. <|start-PI|>A former copper-plate engraver, Longacre had been appointed through<|end-PI|> <|retrieval|> <|start-KFI|>the political influence of South Carolina Senator John C. Calhoun<|end-KFI|>.", "none": null, "results": [ { "Explanation": "The PI introduces the officers at the Philadelphia Mint and sets up a query about the Chief Coiner, while the KFI provides the specific name. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.", "Ideal Knowledge": "A list of officers at the Philadelphia Mint during the mid-19th century, specifically mentioning Franklin Peale as the Chief Coiner.", "Key Factual Information": "Franklin Peale", "Prefix reformulated": "Who was the Chief Coiner at the Philadelphia Mint?", "Prefix-intent": "The officers at the Philadelphia Mint, including Chief Coiner", "Verification": "The KFI directly provides the name of the Chief Coiner mentioned in the PI." }, { "Explanation": "The PI sets up a query about the officers at the Philadelphia Mint, and the KFI provides information about their relationship to Director Patterson. The <|retrieval|> token is inserted to indicate where additional information about the officers' connections might be needed.", "Ideal Knowledge": "Information on the nepotism and personal connections within the Philadelphia Mint during Director Patterson's tenure, including specific examples of friends and family members appointed to positions.", "Key Factual Information": "the friends and relations of Director Patterson", "Prefix reformulated": "What was the relationship between the officers at the Philadelphia Mint and Director Patterson?", "Prefix-intent": "The officers at the Philadelphia Mint were mostly", "Verification": "The KFI directly describes the relationship between the officers and Director Patterson, as prompted by the PI." }, { "Explanation": "The PI introduces the Chief Engraver as an outsider, and the KFI provides his name and additional context. The <|retrieval|> token is inserted to indicate where additional biographical information might be needed.", "Ideal Knowledge": "Biographical information on James B. Longacre, including his appointment as Chief Engraver at the Philadelphia Mint, his predecessor Christian Gobrecht, and the year of Gobrecht's death.", "Key Factual Information": "James B. Longacre, successor to Gobrecht (who had died in 1844)", "Prefix reformulated": "Who was the Chief Engraver at the Philadelphia Mint?", "Prefix-intent": "The outsider in their midst was Chief Engraver", "Verification": "The KFI directly provides the name of the Chief Engraver and additional context, as prompted by the PI." }, { "Explanation": "The PI sets up a query about Longacre's appointment, and the KFI provides information about the political influence behind it. The <|retrieval|> token is inserted to indicate where additional information about the appointment process might be needed.", "Ideal Knowledge": "Information on the political appointment process for positions at the U.S. Mint in the mid-19th century, specifically mentioning Senator John C. Calhoun's role in James B. Longacre's appointment.", "Key Factual Information": "the political influence of South Carolina Senator John C. Calhoun", "Prefix reformulated": "How was James B. Longacre appointed to his position at the Philadelphia Mint?", "Prefix-intent": "A former copper-plate engraver, Longacre had been appointed through", "Verification": "The KFI directly explains how Longacre was appointed, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 Longacre began work on the two new coins in early 1849 , he had no one to assist him . Longacre wrote the following year that he had been warned by a Mint employee that one of the officers ( undoubtedly Peale ) planned to undermine the chief engraver 's position by having the work of preparing designs and dies done outside Mint premises . Accordingly , when the gold coin bill became law , Longacre apprised Patterson that he was ready to begin work on the gold dollar . The Mint Director agreed , and after viewing a model of the head on the obverse , authorized Longacre to proceed with preparation of dies . According to Longacre , \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 Longacre began work on the two new coins in early 1849, he had<|end-PI|> <|retrieval|> <|start-KFI|>no one to assist him<|end-KFI|>. <|start-PI|>Longacre wrote the following year that he had been warned by a Mint employee that one of the officers (undoubtedly Peale) planned to<|end-PI|> <|retrieval|> <|start-KFI|>undermine the chief engraver's position by having the work of preparing designs and dies done outside Mint premises<|end-KFI|>. <|start-PI|>Accordingly, when the gold coin bill became law, Longacre apprised Patterson that<|end-PI|> <|retrieval|> <|start-KFI|>he was ready to begin work on the gold dollar<|end-KFI|>. <|start-PI|>The Mint Director agreed, and after viewing a model of the head on the obverse,<|end-PI|> <|retrieval|> <|start-KFI|>authorized Longacre to proceed with preparation of dies<|end-KFI|>. According to Longacre,", "none": null, "results": [ { "Explanation": "The PI sets up the context of Longacre's work, and the KFI provides the crucial information about his lack of assistance. The <|retrieval|> token is inserted to indicate where additional information about Longacre's working conditions might be needed.", "Ideal Knowledge": "Information about Longacre's work environment and team composition at the U.S. Mint in 1849, specifically detailing his lack of assistants for the new coin designs.", "Key Factual Information": "no one to assist him", "Prefix reformulated": "Who assisted Longacre when he began work on the two new coins in early 1849?", "Prefix-intent": "When Longacre began work on the two new coins in early 1849, he had", "Verification": "The KFI directly answers the question posed by the reformulated PI about Longacre's assistance." }, { "Explanation": "The PI introduces the warning Longacre received, and the KFI provides the specific details of the planned undermining. The <|retrieval|> token is inserted to indicate where additional information about the Mint's internal conflicts might be needed.", "Ideal Knowledge": "Detailed account of the internal politics and conflicts at the U.S. Mint during Longacre's tenure, specifically focusing on attempts to undermine his position as chief engraver.", "Key Factual Information": "undermine the chief engraver's position by having the work of preparing designs and dies done outside Mint premises", "Prefix reformulated": "What did a Mint employee warn Longacre about regarding one of the officers (Peale)?", "Prefix-intent": "Longacre wrote the following year that he had been warned by a Mint employee that one of the officers (undoubtedly Peale) planned to", "Verification": "The KFI directly provides the information about the warning Longacre received, as prompted by the PI." }, { "Explanation": "The PI sets up the context of the gold coin bill becoming law, and the KFI provides Longacre's specific action in response. The <|retrieval|> token is inserted to indicate where additional information about the gold coin bill and Longacre's responsibilities might be needed.", "Ideal Knowledge": "Information about the passage of the gold coin bill and Longacre's immediate actions in response, including his communication with Patterson about starting work on the gold dollar design.", "Key Factual Information": "he was ready to begin work on the gold dollar", "Prefix reformulated": "What did Longacre inform Patterson about when the gold coin bill became law?", "Prefix-intent": "Accordingly, when the gold coin bill became law, Longacre apprised Patterson that", "Verification": "The KFI directly provides Longacre's response to the gold coin bill becoming law, as prompted by the PI." }, { "Explanation": "The PI sets up the context of the Mint Director's review, and the KFI provides the specific authorization given to Longacre. The <|retrieval|> token is inserted to indicate where additional information about the coin design approval process might be needed.", "Ideal Knowledge": "Detailed information about the coin design approval process at the U.S. Mint, specifically focusing on the steps from initial model presentation to die preparation authorization.", "Key Factual Information": "authorized Longacre to proceed with preparation of dies", "Prefix reformulated": "What did the Mint Director authorize Longacre to do after viewing a model of the head on the obverse?", "Prefix-intent": "The Mint Director agreed, and after viewing a model of the head on the obverse,", "Verification": "The KFI directly provides the Mint Director's authorization, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 engraving was unusually minute and required very close and incessant labor for several weeks . I made the original dies and hubs for making the working dies twice over , to secure their perfect adaptation to the coining machinery . I had a wish to execute this work single handed , that I might thus silently reply to those who had questioned my ability for the work . The result , I believe , was satisfactory . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 Type 1 gold dollar depicts a head of Liberty , facing left , with a coronet or tiara on her head bearing her name . Her hair is gathered in a bun ; she is surrounded by 13 stars representing the original states . The reverse features the date and denomination within a wreath , with the name of the nation near the rim . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Contemporary reviews of the Type 1 design were generally favorable . The New York Weekly Tribune on May 19 , 1849 described the new dollar as \" undoubtedly the neatest , tiniest , lightest , coin in this country ... it is too delicate and beautiful to pay out for potatoes , and sauerkraut , and salt pork . Oberon might have paid Puck with it for bringing the blossom which bewitched Titania . \" Willis ' Bank Note List stated that \" there is no probability of them ever getting into general circulation ; they are altogether too small . \" The North Carolina Standard hoped that they would be struck at the Charlotte Mint and circulated locally to eliminate the problem of small @-@ denomination bank notes from out of state . Coin dealer and numismatic author Q. David Bowers notes that the head of Liberty on the Type 1 dollar is a scaled @-@ down version of that on the double eagle , and \" a nicely preserved gold dollar is beautiful to behold \" . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Mint records indicate the first gold dollars were produced on May 7 , 1849 ; Longacre 's diary notes state instead that the first were struck on May 8 . A few coins in proof condition were struck on the first day , along with about 1 @,@ 000 for circulation . There are five major varieties of the 1849 gold dollar from Philadelphia , made as Longacre continued to fine @-@ tune the design . Mintmarked dies were sent by Longacre 's Engraving Department at the Philadelphia Mint to the branch mints at Charlotte , Dahlonega ( in Georgia ) , and New Orleans ; coins struck at the branches resemble some of the types issued from Philadelphia , depending on when the dies were produced . Of the coins struck at the branch mints in 1849 , only pieces struck at Charlotte ( 1849 @-@ C ) exist in multiple varieties ; most are of what is dubbed the \" Closed Wreath \" variety . Approximately five of the 1849 @-@ C Open Wreath are known ; one , believed the finest surviving specimen , sold at auction for $ 690 @,@ 000 in 2004 , remaining a record for the gold dollar series as of 2013 . One of the changes made during production was the inclusion of Longacre 's initial \" L \" on the truncation of Liberty 's neck , the first time a U.S. coin intended for full @-@ scale production had borne the initial of its designer . All issues beginning in 1850 bear the Closed Wreath . Beginning in 1854 , the gold dollar was also struck at the new San Francisco Mint . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 continued flow of gold from California made silver expensive in terms of gold , and U.S. silver coins began to flow out of the country for melting in 1849 , a flow that accelerated over the next several years as the price of the metal continued to rise . By 1853 , a thousand dollars in silver coin contained $ 1 @,@ 042 worth of bullion . As silver coins vanished , the gold dollar became the only federal coin in circulation between the cent and the quarter eagle ( $ 2 @.@ 50 piece ) . As such , it was struck in large numbers and widely circulated . According to Bowers in his book on the denomination , \" the years 1850 to 1853 were the high @-@ water mark of the gold dollar , the glory years of the denomination when the little gold coins took the place of half dollars and silver dollars in everyday transactions . \" This time came to an end in 1853 when Congress passed an act reducing the weight of most silver coins , allowing new issues of them to circulate . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n As early as 1851 , New York Congressman William Duer alleged that that Patterson had made the gold dollar too small in diameter on purpose to provoke criticism . Patterson retired that year after 16 years in his position , and under his successor , George N. Eckert , annular gold dollar and half dollar patterns were struck . Public Ledger reported that although gold dollars would not be struck in annular form , gold half dollars would be , to help fill the need for change . With the new Pierce administration , Thomas M. Pettit took office as Mint Director on March 31 , 1853 . In April , Treasury Secretary James Guthrie wrote to Pettit that there were complaints that the gold dollar was too small , often lost or mistaken for a small silver coin , and enquiring about reports the Mint had experimented with annular dollars . Pettit replied , stating that none had been preserved , but enclosed a silver piece of equivalent size . He noted that while there would be technical difficulties in the production of the annular dollar , these could be overcome . In a letter dated May 10 , Pettit proposed an oval @-@ shaped holed piece , or an angular @-@ shaped coin , which would lessen the production problems . Pettit died suddenly on May 31 ; Guthrie did not let the issue fall , but queried Pettit 's replacement , James Ross Snowden , concerning the issue on June 7 . As U.S. coins were required to bear some device emblematic of liberty , the secretary hoped that artists could be found who could find some such design for an annular coin . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 Act of February 21 , 1853 , that had lightened the silver coins also authorized a gold three @-@ dollar piece , which began to be produced in 1854 . To ensure that the three @-@ dollar piece was not mistaken for other gold coins , it had been made thinner and wider than it would normally be , and Longacre put a distinctive design with an Indian princess on it . Longacre adapted both the technique and the design for the gold dollar , which was made thinner , and thus wider . An adaptation of Longacre 's princess for the larger gold coin was placed on the dollar , and a similar agricultural wreath on the reverse . The idea of making the gold dollar larger in this way had been suggested in Congress as early as 1852 , and had been advocated by Pettit , but Guthrie 's desire for an annular coin stalled the matter . In May 1854 , Snowden sent Guthrie a letter stating that the difficulties with an annular coin , especially in getting the coins to eject properly from the press , were more than trivial . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Nevertheless , the Type 2 gold dollar ( as it came to be known ) proved unsatisfactory as the mints had difficulty in striking the new coin so that all details were brought out . This was due to the high relief of the design — the three Southern branch mints especially had trouble with the piece . Many of the Type 2 pieces quickly became illegible , and were sent back to Philadelphia for melting and recoinage . On most surviving specimens , the \" 85 \" in the date is not fully detailed . The Type 2 gold dollar was struck only at Philadelphia in 1854 and 1855 , at the three Southern branch mints in the latter year , and at San Francisco in 1856 , after the design was designated for replacement . To correct the problems , Longacre enlarged the head of Liberty , making it a scaled @-@ down version of the three @-@ dollar piece , and moved the lettering on the obverse closer to the rim . This improved the metal flow and design sharpness so much that early numismatic scholars assumed the reverse was also altered , though in fact no change was made and the Type 2 and Type 3 reverses are identical . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 Type 2 and 3 gold dollars depict Liberty as a Native American princess , with a fanciful feathered headdress not resembling any worn by any Indian tribe . This image is an inexact copy of the design Longacre had made for the three @-@ dollar piece , and is one of a number of versions of Liberty Longacre created based on the Venus Accroupie or Crouching Venus , a sculpture then on display in a Philadelphia museum . For the reverse , Longacre adapted the \" agricultural wreath \" he had created for the reverse of the three @-@ dollar piece , composed of cotton , corn , tobacco , and wheat , blending the produce of North and South . This wreath would appear , later in the 1850s , on the Flying Eagle cent . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 Type 2 and 3 gold dollars depict Liberty as<|end-PI|> <|retrieval|> <|start-KFI|>a Native American princess, with a fanciful feathered headdress not resembling any worn by any Indian tribe<|end-KFI|>. <|start-PI|>This image is an inexact copy of the design Longacre had made for<|end-PI|> <|retrieval|> <|start-KFI|>the three-dollar piece<|end-KFI|>, and is one of a number of versions of Liberty Longacre created based on the Venus Accroupie or Crouching Venus, a sculpture then on display in a Philadelphia museum. <|start-PI|>For the reverse, Longacre adapted the \"agricultural wreath\" he had created for the reverse of<|end-PI|> <|retrieval|> <|start-KFI|>the three-dollar piece, composed of cotton, corn, tobacco, and wheat, blending the produce of North and South<|end-KFI|>. This wreath would appear, later in the 1850s, on the Flying Eagle cent.", "none": null, "results": [ { "Explanation": "The PI introduces the subject of the Type 2 and 3 gold dollars and sets up a query about their depiction of Liberty. The KFI provides the specific description of the image. The <|retrieval|> token is inserted to indicate where additional information about the coin design might be needed.", "Ideal Knowledge": "Detailed information about the design elements of Type 2 and 3 gold dollars, including the representation of Liberty and the historical context of using Native American imagery in coin designs.", "Key Factual Information": "a Native American princess, with a fanciful feathered headdress not resembling any worn by any Indian tribe", "Prefix reformulated": "How do the Type 2 and 3 gold dollars depict Liberty?", "Prefix-intent": "The Type 2 and 3 gold dollars depict Liberty as", "Verification": "The KFI directly describes how Liberty is depicted on the gold dollars, as prompted by the PI." }, { "Explanation": "The PI sets up a query about the origin of the design, and the KFI provides the specific coin that it was based on. The <|retrieval|> token is inserted to indicate where additional information about Longacre's designs might be needed.", "Ideal Knowledge": "Information about James Longacre's coin designs, specifically the similarities and differences between his designs for the gold dollar and the three-dollar piece.", "Key Factual Information": "the three-dollar piece", "Prefix reformulated": "What coin design is the image on the Type 2 and 3 gold dollars based on?", "Prefix-intent": "This image is an inexact copy of the design Longacre had made for", "Verification": "The KFI directly identifies the coin design that the image is based on, as prompted by the PI." }, { "Explanation": "The PI introduces the subject of the reverse design and sets up a query about its origin. The KFI provides the specific coin it was created for and details about its composition. The <|retrieval|> token is inserted to indicate where additional information about the wreath design might be needed.", "Ideal Knowledge": "Detailed information about Longacre's \"agricultural wreath\" design, including its symbolism, composition, and use on various coins.", "Key Factual Information": "the three-dollar piece, composed of cotton, corn, tobacco, and wheat, blending the produce of North and South", "Prefix reformulated": "What coin did Longacre originally create the \"agricultural wreath\" for?", "Prefix-intent": "For the reverse, Longacre adapted the \"agricultural wreath\" he had created for the reverse of", "Verification": "The KFI directly identifies the coin for which the wreath was originally created and describes its composition, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Art historian Cornelius Vermeule deprecated the Indian princess design used by Longacre for the obverses of the Types 2 and 3 gold dollar , and for the three @-@ dollar piece , \" the ' princess ' of the gold coins is a banknote engraver 's elegant version of folk art of the 1850s . The plumes or feathers are more like the crest of the Prince of Wales than anything that saw the Western frontiers , save perhaps on a music hall beauty . \" \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 gold dollar continued to be produced in the late 1850s , though mintages declined from the figures of two million or more each year between 1850 and 1854 . Only about 51 @,@ 000 gold dollars were produced in 1860 , with over two @-@ thirds of that figure at Philadelphia , just under a third at San Francisco , and 1 @,@ 566 at Dahlonega . Roughly a hundred are known of the last , creating one of the great rarities from Dahlonega in the series . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 other candidate for the rarest from that mint is the 1861 @-@ D , with an estimated mintage of 1 @,@ 000 and perhaps 45 to 60 known . Two pairs of dies were shipped from Philadelphia to Dahlonega on December 10 , 1860 ; they arrived on January 7 , 1861 , two weeks before Georgia voted to secede from the Union , as the American Civil War began . Under orders from Governor Joseph E. Brown , state militia secured the mint , and at some point , small quantities of dollars and half eagles were produced . Records of how many coins were struck and when have not survived . Since dies crack in time , and all the mints were supplied with them from Philadelphia , coining could not last , and in May 1861 , coins and supplies remaining at Dahlonega were turned over to the treasury of the Confederate States of America , which Georgia had by then joined . Gold coins with a face value of $ 6 were put aside for assay . Normally , they would have been sent to Philadelphia to await the following year 's meeting of the United States Assay Commission , when they would be available for testing . Instead , these were sent to the initial Confederate capital of Montgomery , Alabama , though what was done with them there , and their ultimate fate , are unknown . The rarity of the 1861 @-@ D dollar , and the association with the Confederacy , make it especially prized . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 other candidate for the rarest from that mint is the 1861 @-@ D , with an estimated mintage of<|end-PI|> <|retrieval|> <|start-KFI|>1 @,@ 000 and perhaps 45 to 60 known<|end-KFI|> . <|start-PI|>Two pairs of dies were shipped from Philadelphia to Dahlonega on<|end-PI|> <|retrieval|> <|start-KFI|>December 10 , 1860 ; they arrived on January 7 , 1861 , two weeks before Georgia voted to secede from the Union , as the American Civil War began<|end-KFI|> . Under orders from Governor Joseph E. Brown , state militia secured the mint , and at some point , small quantities of dollars and half eagles were produced . Records of how many coins were struck and when have not survived . Since dies crack in time , and all the mints were supplied with them from Philadelphia , coining could not last , and in May 1861 , coins and supplies remaining at Dahlonega were turned over to the treasury of the Confederate States of America , which Georgia had by then joined . <|start-PI|>Gold coins with a face value of $ 6 were put aside for assay . Normally , they would have been sent to Philadelphia to await<|end-PI|> <|retrieval|> <|start-KFI|>the following year 's meeting of the United States Assay Commission , when they would be available for testing<|end-KFI|> . Instead , these were sent to the initial Confederate capital of Montgomery , Alabama , though what was done with them there , and their ultimate fate , are unknown . <|start-PI|>The rarity of the 1861 @-@ D dollar , and the association with the Confederacy , make it<|end-PI|> <|retrieval|> <|start-KFI|>especially prized<|end-KFI|> .", "none": null, "results": [ { "Explanation": "The PI introduces the 1861-D coin as a rare candidate and sets up a query about its mintage, while the KFI provides the specific numbers. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.", "Ideal Knowledge": "Detailed information about rare coins from the Dahlonega mint, including mintage numbers and survival rates for the 1861-D coin.", "Key Factual Information": "1 @,@ 000 and perhaps 45 to 60 known", "Prefix reformulated": "What is the estimated mintage of the 1861-D coin?", "Prefix-intent": "The other candidate for the rarest from that mint is the 1861 @-@ D , with an estimated mintage of", "Verification": "The KFI directly provides the estimated mintage and known quantity of the 1861-D coin, as prompted by the PI." }, { "Explanation": "The PI sets up a query about the shipment of dies, and the KFI provides specific dates and historical context. The <|retrieval|> token is inserted to indicate where additional information about the historical timeline might be needed.", "Ideal Knowledge": "A timeline of events leading up to the American Civil War, focusing on the movement of minting equipment and the secession of Southern states, particularly Georgia.", "Key Factual Information": "December 10 , 1860 ; they arrived on January 7 , 1861 , two weeks before Georgia voted to secede from the Union , as the American Civil War began", "Prefix reformulated": "When were two pairs of dies shipped from Philadelphia to Dahlonega?", "Prefix-intent": "Two pairs of dies were shipped from Philadelphia to Dahlonega on", "Verification": "The KFI directly provides the shipping and arrival dates of the dies, as well as contextual information about Georgia's secession, as prompted by the PI." }, { "Explanation": "The PI introduces the topic of gold coins set aside for assay and sets up a query about the normal procedure, while the KFI provides the specific details. The <|retrieval|> token is inserted to indicate where additional information about the Assay Commission might be needed.", "Ideal Knowledge": "Information about the United States Assay Commission's procedures, including the annual meeting and testing process for newly minted coins.", "Key Factual Information": "the following year 's meeting of the United States Assay Commission , when they would be available for testing", "Prefix reformulated": "What would normally happen to gold coins put aside for assay?", "Prefix-intent": "Gold coins with a face value of $ 6 were put aside for assay . Normally , they would have been sent to Philadelphia to await", "Verification": "The KFI directly explains the normal procedure for coins set aside for assay, as prompted by the PI." }, { "Explanation": "The PI sets up the reasons for the coin's value, and the KFI confirms its prized status. The <|retrieval|> token is inserted to indicate where additional information about coin valuation and historical significance might be needed.", "Ideal Knowledge": "Detailed information about the factors that contribute to the value of rare coins, particularly those associated with historical events like the Civil War and the Confederacy.", "Key Factual Information": "especially prized", "Prefix reformulated": "Why is the 1861-D dollar especially prized?", "Prefix-intent": "The rarity of the 1861 @-@ D dollar , and the association with the Confederacy , make it", "Verification": "The KFI directly states that the coin is especially prized, as prompted by the PI which sets up the reasons for its value." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Dahlonega , like the other two branch mints in the South , closed its doors after the 1861 strikings . It and the Charlotte facility never reopened ; the New Orleans Mint again struck coins from 1879 to 1909 , but did not strike gold dollars again . After 1861 , the only issuance of gold dollars outside Philadelphia was at San Francisco , in 1870 . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 outbreak of the Civil War shook public confidence in the Union , and citizens began hoarding specie , gold and silver coins . In late December 1861 , banks and then the federal Treasury stopped paying out gold at face value . By mid @-@ 1862 , all federal coins , even the base metal cent , had vanished from commerce in much of the country . The exception was the Far West , where for the most part , only gold and silver were acceptable currencies , and paper money traded at a discount . In the rest of the nation , gold and silver coins could be purchased from banks , exchange agents , and from the Treasury for a premium in the new greenbacks the government began to issue to fill the gap in commerce and finance the war . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.", "role": "user" } ]

{ "Updated_Passage": "<|start-PI|>The outbreak of the Civil War shook public confidence in the Union, and citizens began<|end-PI|> <|retrieval|> <|start-KFI|>hoarding specie, gold and silver coins<|end-KFI|>. <|start-PI|>In late December 1861, banks and then the federal Treasury stopped<|end-PI|> <|retrieval|> <|start-KFI|>paying out gold at face value<|end-KFI|>. <|start-PI|>By mid @-@ 1862, all federal coins, even the base metal cent, had vanished from commerce in<|end-PI|> <|retrieval|> <|start-KFI|>much of the country<|end-KFI|>. <|start-PI|>The exception was the Far West, where for the most part, only gold and silver were acceptable currencies, and<|end-PI|> <|retrieval|> <|start-KFI|>paper money traded at a discount<|end-KFI|>. <|start-PI|>In the rest of the nation, gold and silver coins could be purchased from banks, exchange agents, and from the Treasury for<|end-PI|> <|retrieval|> <|start-KFI|>a premium in the new greenbacks the government began to issue to fill the gap in commerce and finance the war<|end-KFI|>.", "none": null, "results": [ { "Explanation": "The PI sets up the context of the Civil War and its impact on public confidence, while the KFI provides the specific action taken by citizens. The <|retrieval|> token is inserted to indicate where additional information about citizen behavior during wartime might be needed.", "Ideal Knowledge": "Information on economic behaviors during times of war, specifically detailing how citizens tend to hoard precious metals and coins when confidence in the government is shaken.", "Key Factual Information": "hoarding specie, gold and silver coins", "Prefix reformulated": "What did citizens begin doing when the Civil War shook public confidence in the Union?", "Prefix-intent": "The outbreak of the Civil War shook public confidence in the Union, and citizens began", "Verification": "The KFI directly provides the action taken by citizens as prompted by the PI." }, { "Explanation": "The PI provides the time frame and the entities involved, while the KFI specifies the action that was stopped. The <|retrieval|> token is inserted to indicate where additional information about financial policies during the Civil War might be needed.", "Ideal Knowledge": "Detailed information on the financial policies and actions taken by banks and the federal Treasury during the Civil War, specifically focusing on the suspension of gold payments at face value in late 1861.", "Key Factual Information": "paying out gold at face value", "Prefix reformulated": "What did banks and the federal Treasury stop doing in late December 1861?", "Prefix-intent": "In late December 1861, banks and then the federal Treasury stopped", "Verification": "The KFI directly answers what banks and the federal Treasury stopped doing, as prompted by the PI." }, { "Explanation": "The PI sets up the context of the coin shortage by mid-1862, while the KFI specifies the geographical extent. The <|retrieval|> token is inserted to indicate where additional information about the spread of the coin shortage might be needed.", "Ideal Knowledge": "A comprehensive overview of the coin shortage during the Civil War, including its geographical extent and impact on different regions of the country.", "Key Factual Information": "much of the country", "Prefix reformulated": "Where had all federal coins vanished from commerce by mid-1862?", "Prefix-intent": "By mid @-@ 1862, all federal coins, even the base metal cent, had vanished from commerce in", "Verification": "The KFI directly provides the extent of the coin shortage, as prompted by the PI." }, { "Explanation": "The PI sets up the context of the Far West's unique monetary situation, while the KFI specifies how paper money was treated. The <|retrieval|> token is inserted to indicate where additional information about regional monetary differences during the Civil War might be needed.", "Ideal Knowledge": "Detailed information on the monetary practices in the Far West during the Civil War, focusing on the preference for gold and silver and the discounting of paper money.", "Key Factual Information": "paper money traded at a discount", "Prefix reformulated": "How was paper money treated in the Far West during the Civil War?", "Prefix-intent": "The exception was the Far West, where for the most part, only gold and silver were acceptable currencies, and", "Verification": "The KFI directly provides information about the treatment of paper money in the Far West, as prompted by the PI." }, { "Explanation": "The PI sets up the context of purchasing gold and silver coins, while the KFI specifies the method of purchase and the purpose of the new currency. The <|retrieval|> token is inserted to indicate where additional information about the use of greenbacks during the Civil War might be needed.", "Ideal Knowledge": "Information on the issuance of greenbacks during the Civil War, their use in purchasing gold and silver coins, and their role in financing the war effort.", "Key Factual Information": "a premium in the new greenbacks the government began to issue to fill the gap in commerce and finance the war", "Prefix reformulated": "How could gold and silver coins be purchased in most of the nation during the Civil War?", "Prefix-intent": "In the rest of the nation, gold and silver coins could be purchased from banks, exchange agents, and from the Treasury for", "Verification": "The KFI directly provides information about how gold and silver coins could be purchased, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Since gold did not circulate in the United States ( except on the West Coast ) in the postwar period , much of the production of coins of that metal in the United States was double eagles for export . Accordingly , although 1 @,@ 361 @,@ 355 gold dollars were struck in 1862 — the last time production would exceed a million — the mintage fell to 6 @,@ 200 in 1863 and remained low for the rest of the coin 's existence , excepting 1873 and 1874 . The Mint felt it improper to suspend coinage of a coin authorized by Congress , and issued proof coins ( generally a few dozen to the tiny numismatic community ) from specially @-@ polished dies , also producing enough circulation strikes so that the proof coins would not be unduly rare . In 1873 and 1874 , old and worn gold dollars held by the government were melted and recoined , generating large mintages of that denomination . This was done in anticipation of the resumption of specie payments , which did not occur until the end of 1878 . Once specie again circulated at face value , the gold dollar found no place in commerce amid large quantities of silver coinage , either released from hoarding or newly struck by the Mint . The government expected that the resumption of specie payments would cause the dollar and other small gold coins to circulate again , but the public , allowed to redeem paper currency , continued to use it as more convenient than coins . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 1870s and 1880s , public interest grew in the low @-@ mintage gold dollar . Collecting coins was becoming more popular , and a number of numismatists put aside some gold dollars and hoped for increases in value . The Mint most likely channeled its production through some favored Philadelphia dealers , though proof coins could be purchased for $ 1 @.@ 25 at the cashier 's window at the Philadelphia facility . Banks charged a premium for circulation strikes . They were popular in the jewelry trade , mounted into various items . The coins were often exported to China or Japan , where such jewelry was made . The dollars were often damaged in the process ; the Mint refused to sell into this trade and did its best to hinder it . Nevertheless , Mint officials concluded that jewelers were successful at getting the majority of each issue . Proof mintages exceeded 1 @,@ 000 by 1884 , and remained above that mark for the remainder of the series , numbers likely inflated by agents of jewelers , willing to pay the Mint 's premium of $ .25 per coin . Another use for the gold dollar was as a holiday gift ; after its abolition the quarter eagle became a popular present . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n James Pollock , in his final report as Mint Director in 1873 , advocated limiting striking of gold dollars to depositors who specifically requested it . \" The gold dollar is not a convenient coin , on account of its small size , and it suffers more proportionately from abrasion than larger coins . \" His successors called for its abolition , with James P. Kimball , before he left office in 1889 , writing to Congress that except as jewelry , \" little practical use has been found for this coin \" . Later that year , the new director , Edward O. Leech , issued a report stating that the gold dollar \" is too small for circulation , and ... [ is ] used almost exclusively for the purposes of ornament . The last year in which the gold dollar was struck was 1889 . Congress abolished the gold dollar , along with the three @-@ cent nickel and three @-@ dollar piece , by the Act of September 26 , 1890 . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 total of 19 @,@ 499 @,@ 337 gold dollars were coined , of which 18 @,@ 223 @,@ 438 were struck at Philadelphia , 1 @,@ 004 @,@ 000 at New Orleans , 109 @,@ 138 at Charlotte , 90 @,@ 232 at San Francisco and 72 @,@ 529 at Dahlonega . According to an advertisement in the February 1899 issue of The Numismatist , gold dollars brought $ 1 @.@ 80 each , still in demand as a birthday present and for jewelry . That journal in 1905 carried news of a customer depositing 100 gold dollars into a bank ; the teller , aware of the value , credited the account with $ 1 @.@ 60 per coin . In 1908 , a dealer offered $ 2 each for any quantity . As coin collecting became a widespread pastime in the early 20th century , gold dollars became a popular specialty , a status they retain . The 2014 edition of R.S. Yeoman 's A Guide Book of United States Coins rates the least expensive gold dollar in very fine condition ( VF @-@ 20 ) at $ 300 , a value given for each of the Type 1 Philadelphia issues from 1849 to 1853 . Those seeking one of each type will find the most expensive to be a specimen of the Type 2 , with the 1854 and 1855 estimated at $ 350 in that condition ; the other two types have dates valued at $ 300 in that grade . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 total of 19,499,337 gold dollars were coined, of which<|end-PI|> <|retrieval|> <|start-KFI|>18,223,438 were struck at Philadelphia, 1,004,000 at New Orleans, 109,138 at Charlotte, 90,232 at San Francisco and 72,529 at Dahlonega<|end-KFI|>. <|start-PI|>According to an advertisement in the February 1899 issue of The Numismatist, gold dollars brought<|end-PI|> <|retrieval|> <|start-KFI|>$ 1.80 each, still in demand as a birthday present and for jewelry<|end-KFI|>. That journal in 1905 carried news of a customer depositing 100 gold dollars into a bank; the teller, aware of the value, credited the account with $ 1.60 per coin. In 1908, a dealer offered $ 2 each for any quantity. As coin collecting became a widespread pastime in the early 20th century, gold dollars became a popular specialty, a status they retain. <|start-PI|>The 2014 edition of R.S. Yeoman's A Guide Book of United States Coins rates the least expensive gold dollar in very fine condition (VF-20) at<|end-PI|> <|retrieval|> <|start-KFI|>$ 300, a value given for each of the Type 1 Philadelphia issues from 1849 to 1853<|end-KFI|>. Those seeking one of each type will find the most expensive to be a specimen of the Type 2, with the 1854 and 1855 estimated at $ 350 in that condition; the other two types have dates valued at $ 300 in that grade.", "none": null, "results": [ { "Explanation": "The PI introduces the total number of gold dollars coined and sets up a query about their distribution, while the KFI provides the specific numbers for each mint. The <|retrieval|> token is inserted to indicate where additional information about mint production might be needed.", "Ideal Knowledge": "A detailed breakdown of gold dollar production by mint, including historical context for each mint's operations and their relative importance in gold coin production.", "Key Factual Information": "18,223,438 were struck at Philadelphia, 1,004,000 at New Orleans, 109,138 at Charlotte, 90,232 at San Francisco and 72,529 at Dahlonega", "Prefix reformulated": "How many gold dollars were coined at different mints?", "Prefix-intent": "A total of 19,499,337 gold dollars were coined, of which", "Verification": "The KFI directly provides the breakdown of gold dollars coined at different mints, as prompted by the PI." }, { "Explanation": "The PI sets up a query about the price of gold dollars in 1899, and the KFI provides the specific price and additional context. The <|retrieval|> token is inserted to indicate where additional information about historical pricing might be needed.", "Ideal Knowledge": "Historical pricing data for gold dollars in the late 19th century, including information on their use in jewelry and as gifts, and how this affected their numismatic value.", "Key Factual Information": "$ 1.80 each, still in demand as a birthday present and for jewelry", "Prefix reformulated": "What was the price of gold dollars according to an 1899 advertisement?", "Prefix-intent": "According to an advertisement in the February 1899 issue of The Numismatist, gold dollars brought", "Verification": "The KFI directly provides the price and demand information for gold dollars, as prompted by the PI." }, { "Explanation": "The PI introduces a specific coin guide and sets up a query about the value of certain gold dollars, while the KFI provides the exact value and additional context. The <|retrieval|> token is inserted to indicate where additional information about coin grading and valuation might be needed.", "Ideal Knowledge": "Detailed pricing information for gold dollars in various conditions, with specific focus on Type 1 Philadelphia issues from 1849 to 1853, including factors that influence their value.", "Key Factual Information": "$ 300, a value given for each of the Type 1 Philadelphia issues from 1849 to 1853", "Prefix reformulated": "What is the value of the least expensive gold dollar in very fine condition according to the 2014 R.S. Yeoman's guide?", "Prefix-intent": "The 2014 edition of R.S. Yeoman's A Guide Book of United States Coins rates the least expensive gold dollar in very fine condition (VF-20) at", "Verification": "The KFI directly provides the value and specific information about the coins with this value, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 gold dollar had a brief resurrection during the period of Early United States commemorative coins . Between 1903 and 1922 nine different issues were produced , with a total mintage of 99 @,@ 799 . These were minted for various public events , did not circulate , and none used Longacre 's design . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 Johnson – Corey – Chaykovsky reaction ( sometimes referred to as the Corey – Chaykovsky reaction or CCR ) is a chemical reaction used in organic chemistry for the synthesis of epoxides , aziridines , and cyclopropanes . It was discovered in 1961 by A. William Johnson and developed significantly by E. J. Corey and Michael Chaykovsky . The reaction involves addition of a sulfur ylide to a ketone , aldehyde , imine , or enone to produce the corresponding 3 @-@ membered ring . The reaction is diastereoselective favoring trans substitution in the product regardless of the initial stereochemistry . The synthesis of epoxides via this method serves as an important retrosynthetic alternative to the traditional epoxidation reactions of olefins . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 reaction is most often employed for epoxidation via methylene transfer , and to this end has been used in several notable total syntheses ( See Synthesis of epoxides below ) . Additionally detailed below are the history , mechanism , scope , and enantioselective variants of the reaction . Several reviews have been published . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 original publication by Johnson concerned the reaction of 9 @-@ dimethylsulfonium fluorenylide with substituted benzaldehyde derivatives . The attempted Wittig @-@ like reaction failed and a benzalfluorene oxide was obtained instead , noting that \" Reaction between the sulfur ylid and benzaldehydes did not afford benzalfluorenes as had the phosphorus and arsenic ylids . \" \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 original publication by Johnson concerned the reaction of<|end-PI|> <|retrieval|> <|start-KFI|>9 @-@ dimethylsulfonium fluorenylide with substituted benzaldehyde derivatives<|end-KFI|> . <|start-PI|>The attempted Wittig @-@ like reaction failed and<|end-PI|> <|retrieval|> <|start-KFI|>a benzalfluorene oxide was obtained instead<|end-KFI|> , noting that \" <|start-PI|>Reaction between the sulfur ylid and benzaldehydes did not afford<|end-PI|> <|retrieval|> <|start-KFI|>benzalfluorenes as had the phosphorus and arsenic ylids<|end-KFI|> . \"", "none": null, "results": [ { "Explanation": "The PI introduces Johnson's publication and sets up a query about the reaction studied, while the KFI provides the specific chemical compounds involved. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.", "Ideal Knowledge": "Detailed information about Johnson's publication, including the exact chemical compounds involved in the reaction, their structures, and the context of the study.", "Key Factual Information": "9 @-@ dimethylsulfonium fluorenylide with substituted benzaldehyde derivatives", "Prefix reformulated": "What was the reaction studied in Johnson's original publication?", "Prefix-intent": "The original publication by Johnson concerned the reaction of", "Verification": "The KFI directly provides the specific reaction components mentioned in the PI." }, { "Explanation": "The PI sets up the context of a failed reaction, and the KFI provides the unexpected product obtained. The <|retrieval|> token is inserted to indicate where additional information about the reaction outcome might be needed.", "Ideal Knowledge": "Detailed explanation of the Wittig reaction, its mechanism, expected products, and possible side reactions or unexpected outcomes, specifically focusing on the formation of benzalfluorene oxide.", "Key Factual Information": "a benzalfluorene oxide was obtained instead", "Prefix reformulated": "What was the result of the attempted Wittig-like reaction?", "Prefix-intent": "The attempted Wittig @-@ like reaction failed and", "Verification": "The KFI directly provides the outcome of the failed reaction, as prompted by the PI." }, { "Explanation": "The PI sets up a comparison between different ylid reactions, and the KFI provides the specific product that was not obtained with sulfur ylids, contrasting it with phosphorus and arsenic ylids. The <|retrieval|> token is inserted to indicate where additional information about these comparative reactions might be needed.", "Ideal Knowledge": "Comparative information on reactions involving sulfur, phosphorus, and arsenic ylids with benzaldehydes, focusing on their different products and reaction mechanisms.", "Key Factual Information": "benzalfluorenes as had the phosphorus and arsenic ylids", "Prefix reformulated": "What did the reaction between the sulfur ylid and benzaldehydes not produce?", "Prefix-intent": "Reaction between the sulfur ylid and benzaldehydes did not afford", "Verification": "The KFI directly states what the reaction did not produce, contrasting it with the results of other ylids, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 subsequent development of ( dimethyloxosulfaniumyl ) methanide , ( CH3 ) 2SOCH2 and ( dimethylsulfaniumyl ) methanide , ( CH3 ) 2SCH2 ( known as Corey – Chaykovsky reagents ) by Corey and Chaykovsky as efficient methylene @-@ transfer reagents established the reaction as a part of the organic canon . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 reaction mechanism for the Johnson – Corey – Chaykovsky reaction consists of nucleophilic addition of the ylide to the carbonyl or imine group . A negative charge is transferred to the heteroatom and because the sulfonium cation is a good leaving group it gets expelled forming the ring . In the related Wittig reaction , the formation of the much stronger phosphorus @-@ oxygen double bond prevents oxirane formation and instead , olefination takes place through a 4 @-@ membered cyclic intermediate . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 reaction mechanism for the Johnson – Corey – Chaykovsky reaction consists of<|end-PI|> <|retrieval|> <|start-KFI|>nucleophilic addition of the ylide to the carbonyl or imine group<|end-KFI|>. <|start-PI|>A negative charge is transferred to the heteroatom and because the sulfonium cation is a good leaving group<|end-PI|> <|retrieval|> <|start-KFI|>it gets expelled forming the ring<|end-KFI|>. <|start-PI|>In the related Wittig reaction , the formation of the much stronger phosphorus @-@ oxygen double bond prevents oxirane formation and instead<|end-PI|> <|retrieval|> <|start-KFI|>olefination takes place through a 4 @-@ membered cyclic intermediate<|end-KFI|>.", "none": null, "results": [ { "Explanation": "The PI introduces the reaction mechanism, and the KFI provides the specific process involved. The <|retrieval|> token is inserted to indicate where additional information about the reaction mechanism might be needed.", "Ideal Knowledge": "Detailed explanation of the Johnson – Corey – Chaykovsky reaction mechanism, including the role of ylides and their interaction with carbonyl or imine groups in nucleophilic addition.", "Key Factual Information": "nucleophilic addition of the ylide to the carbonyl or imine group", "Prefix reformulated": "What does the reaction mechanism for the Johnson – Corey – Chaykovsky reaction consist of?", "Prefix-intent": "The reaction mechanism for the Johnson – Corey – Chaykovsky reaction consists of", "Verification": "The KFI directly describes the main component of the reaction mechanism, as prompted by the PI." }, { "Explanation": "The PI sets up the conditions for the next step in the reaction, and the KFI provides the resulting action. The <|retrieval|> token is inserted to indicate where additional information about this specific step might be needed.", "Ideal Knowledge": "Explanation of charge transfer in the Johnson – Corey – Chaykovsky reaction, focusing on the role of the sulfonium cation as a leaving group and the subsequent ring formation process.", "Key Factual Information": "it gets expelled forming the ring", "Prefix reformulated": "What happens when a negative charge is transferred to the heteroatom in the Johnson – Corey – Chaykovsky reaction?", "Prefix-intent": "A negative charge is transferred to the heteroatom and because the sulfonium cation is a good leaving group", "Verification": "The KFI directly describes the consequence of the charge transfer and the role of the sulfonium cation, as prompted by the PI." }, { "Explanation": "The PI sets up the contrast between the two reactions, and the KFI provides the specific process that occurs in the Wittig reaction. The <|retrieval|> token is inserted to indicate where additional information about the Wittig reaction mechanism might be needed.", "Ideal Knowledge": "Detailed comparison of the Johnson – Corey – Chaykovsky and Wittig reactions, focusing on the role of the phosphorus-oxygen double bond in preventing oxirane formation and the mechanism of olefination through a 4-membered cyclic intermediate.", "Key Factual Information": "olefination takes place through a 4 @-@ membered cyclic intermediate", "Prefix reformulated": "What happens in the Wittig reaction instead of oxirane formation?", "Prefix-intent": "In the related Wittig reaction , the formation of the much stronger phosphorus @-@ oxygen double bond prevents oxirane formation and instead", "Verification": "The KFI directly describes the alternative process that occurs in the Wittig reaction, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 trans diastereoselectivity observed results from the reversibility of the initial addition , allowing equilibration to the favored anti betaine over the syn betaine . Initial addition of the ylide results in a betaine with adjacent charges ; density functional theory calculations have shown that the rate @-@ limiting step is rotation of the central bond into the conformer necessary for backside attack on the sulfonium . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 trans diastereoselectivity observed results from the reversibility of the initial addition , allowing equilibration to<|end-PI|> <|retrieval|> <|start-KFI|>the favored anti betaine over the syn betaine<|end-KFI|> . <|start-PI|>Initial addition of the ylide results in a betaine with adjacent charges ; density functional theory calculations have shown that the rate @-@ limiting step is<|end-PI|> <|retrieval|> <|start-KFI|>rotation of the central bond into the conformer necessary for backside attack on the sulfonium<|end-KFI|> .", "none": null, "results": [ { "Explanation": "The PI sets up the context for the trans diastereoselectivity and its cause, while the KFI provides the specific outcome of the equilibration process. The <|retrieval|> token is inserted to indicate where additional information about betaine formation might be needed.", "Ideal Knowledge": "Detailed information about the mechanism of trans diastereoselectivity in chemical reactions, specifically focusing on the equilibration between anti and syn betaines and their relative stabilities.", "Key Factual Information": "the favored anti betaine over the syn betaine", "Prefix reformulated": "What does the trans diastereoselectivity in this reaction result from?", "Prefix-intent": "The trans diastereoselectivity observed results from the reversibility of the initial addition , allowing equilibration to", "Verification": "The KFI directly provides the result of the equilibration process mentioned in the PI." }, { "Explanation": "The PI introduces the initial addition of the ylide and mentions density functional theory calculations, setting up for the revelation of the rate-limiting step. The KFI provides the specific details of this step. The <|retrieval|> token is inserted to indicate where additional information about reaction mechanisms and computational chemistry might be needed.", "Ideal Knowledge": "Detailed explanation of density functional theory calculations applied to ylide addition reactions, focusing on the energetics of bond rotation and its role in determining the rate-limiting step of the reaction.", "Key Factual Information": "rotation of the central bond into the conformer necessary for backside attack on the sulfonium", "Prefix reformulated": "What is the rate-limiting step in the reaction of ylide addition according to density functional theory calculations?", "Prefix-intent": "Initial addition of the ylide results in a betaine with adjacent charges ; density functional theory calculations have shown that the rate @-@ limiting step is", "Verification": "The KFI directly provides the rate-limiting step as determined by the calculations mentioned in the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 of reversibility in the initial step ( and therefore the diastereoselectivity ) depends on four factors , with greater reversibility corresponding to higher selectivity : \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Stability of the substrate with higher stability leading to greater reversibility by favoring the starting material over the betaine . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Steric hindrance in the betaine with greater hindrance leading to greater reversibility by disfavoring formation of the intermediate and slowing the rate @-@ limiting rotation of the central bond . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Solvation of charges in the betaine by counterions such as lithium with greater solvation allowing more facile rotation in the betaine intermediate , lowering the amount of reversibility . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 application of the Johnson – Corey – Chaykovsky reaction in organic synthesis is diverse . The reaction has come to encompass reactions of many types of sulfur ylides with electrophiles well beyond the original publications . It has seen use in a number of high @-@ profile total syntheses , as detailed below , and is generally recognized as a powerful transformative tool in the organic repertoire . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 types of ylides can be prepared with various functional groups both on the anionic carbon center and on the sulfur . The substitution pattern can influence the ease of preparation for the reagents ( typically from the sulfonium halide , e.g. trimethylsulfonium iodide ) and overall reaction rate in various ways . The general format for the reagent is shown on the right . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Use of a sulfoxonium allows more facile preparation of the reagent using weaker bases as compared to sulfonium ylides . ( The difference being that a sulfoxonium contains a doubly bonded oxygen whereas the sulfonium does not . ) The former react slower due to their increased stability . In addition , the dialkylsulfoxide by @-@ products of sulfoxonium reagents are greatly preferred to the significantly more toxic , volatile , and odorous dialkylsulfide by @-@ products from sulfonium reagents . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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|>Use of a sulfoxonium allows more facile preparation of the reagent using<|end-PI|> <|retrieval|> <|start-KFI|>weaker bases as compared to sulfonium ylides<|end-KFI|> . ( <|start-PI|>The difference being that a sulfoxonium contains<|end-PI|> <|retrieval|> <|start-KFI|>a doubly bonded oxygen whereas the sulfonium does not<|end-KFI|> . ) <|start-PI|>The former react slower due to<|end-PI|> <|retrieval|> <|start-KFI|>their increased stability<|end-KFI|> . <|start-PI|>In addition , the dialkylsulfoxide by @-@ products of sulfoxonium reagents are<|end-PI|> <|retrieval|> <|start-KFI|>greatly preferred to the significantly more toxic , volatile , and odorous dialkylsulfide by @-@ products from sulfonium reagents<|end-KFI|> .", "none": null, "results": [ { "Explanation": "The PI introduces the use of sulfoxonium and sets up a query about its advantage, while the KFI provides the specific benefit. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.", "Ideal Knowledge": "A comparison of sulfoxonium and sulfonium ylides in reagent preparation, focusing on the strength of bases required and the ease of preparation.", "Key Factual Information": "weaker bases as compared to sulfonium ylides", "Prefix reformulated": "What does the use of a sulfoxonium allow in reagent preparation?", "Prefix-intent": "Use of a sulfoxonium allows more facile preparation of the reagent using", "Verification": "The KFI directly provides information about the advantage of using sulfoxonium in reagent preparation, as prompted by the PI." }, { "Explanation": "The PI sets up a query about the difference between sulfoxonium and sulfonium, and the KFI provides the specific structural difference. The <|retrieval|> token is inserted to indicate where additional information about chemical structures might be needed.", "Ideal Knowledge": "Detailed structural information about sulfoxonium and sulfonium compounds, highlighting the presence of a doubly bonded oxygen in sulfoxonium.", "Key Factual Information": "a doubly bonded oxygen whereas the sulfonium does not", "Prefix reformulated": "What does a sulfoxonium contain that differentiates it from a sulfonium?", "Prefix-intent": "The difference being that a sulfoxonium contains", "Verification": "The KFI directly provides the structural difference between sulfoxonium and sulfonium, as prompted by the PI." }, { "Explanation": "The PI sets up a query about the reaction speed of sulfoxonium compounds, and the KFI provides the reason for their slower reaction. The <|retrieval|> token is inserted to indicate where additional information about chemical reactivity might be needed.", "Ideal Knowledge": "Information on the reactivity of sulfoxonium compounds compared to sulfonium ylides, explaining how increased stability affects reaction rates.", "Key Factual Information": "their increased stability", "Prefix reformulated": "Why do sulfoxonium compounds react slower?", "Prefix-intent": "The former react slower due to", "Verification": "The KFI directly provides the reason for the slower reaction of sulfoxonium compounds, as prompted by the PI." }, { "Explanation": "The PI introduces the by-products of sulfoxonium reagents and sets up a comparison, while the KFI provides specific advantages over sulfonium reagent by-products. The <|retrieval|> token is inserted to indicate where additional information about chemical by-products and their properties might be needed.", "Ideal Knowledge": "A detailed comparison of the by-products produced by sulfoxonium and sulfonium reagents, focusing on toxicity, volatility, odor, and overall safety considerations.", "Key Factual Information": "greatly preferred to the significantly more toxic , volatile , and odorous dialkylsulfide by @-@ products from sulfonium reagents", "Prefix reformulated": "What are the characteristics of dialkylsulfoxide by-products from sulfoxonium reagents?", "Prefix-intent": "In addition , the dialkylsulfoxide by @-@ products of sulfoxonium reagents are", "Verification": "The KFI directly compares the by-products of sulfoxonium and sulfonium reagents, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 vast majority of reagents are monosubstituted at the ylide carbon ( either R1 or R2 as hydrogen ) . Disubstituted reagents are much rarer but have been described : \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n If the ylide carbon is substituted with an electron @-@ withdrawing group ( EWG ) , the reagent is referred to as a stabilized ylide . These , similarly to sulfoxonium reagents , react much slower and are typically easier to prepare . These are limited in their usefulness as the reaction can become prohibitively sluggish : examples involving amides are widespread , with many fewer involving esters and virtually no examples involving other EWG 's . For these , the related Darzens reaction is typically more appropriate . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.", "role": "user" } ]

{ "Updated_Passage": "If the ylide carbon is substituted with an electron @-@ withdrawing group ( EWG ) , <|start-PI|> the reagent is referred to as <|end-PI|> <|retrieval|> <|start-KFI|> a stabilized ylide <|end-KFI|> . <|start-PI|> These , similarly to sulfoxonium reagents , react <|end-PI|> <|retrieval|> <|start-KFI|> much slower and are typically easier to prepare <|end-KFI|> . <|start-PI|> These are limited in their usefulness as the reaction can become <|end-PI|> <|retrieval|> <|start-KFI|> prohibitively sluggish : examples involving amides are widespread , with many fewer involving esters and virtually no examples involving other EWG 's <|end-KFI|> . <|start-PI|> For these , the related <|end-PI|> <|retrieval|> <|start-KFI|> Darzens reaction is typically more appropriate <|end-KFI|> .", "none": null, "results": [ { "Explanation": "The PI sets up the context for a specific type of ylide, and the KFI provides the name for this type. The <|retrieval|> token is inserted to indicate where additional information about ylide classification might be needed.", "Ideal Knowledge": "A detailed explanation of ylide chemistry, focusing on the effects of electron-withdrawing groups on ylide stability and reactivity, with specific mention of stabilized ylides.", "Key Factual Information": "a stabilized ylide", "Prefix reformulated": "What is the reagent called when the ylide carbon is substituted with an electron-withdrawing group (EWG)?", "Prefix-intent": "If the ylide carbon is substituted with an electron @-@ withdrawing group ( EWG ) , the reagent is referred to as", "Verification": "The KFI directly provides the name of the reagent as requested in the PI." }, { "Explanation": "The PI compares stabilized ylides to sulfoxonium reagents, and the KFI provides specific information about their reactivity and preparation. The <|retrieval|> token is inserted to indicate where additional comparative information might be needed.", "Ideal Knowledge": "Comparative analysis of the reactivity and preparation methods of stabilized ylides and sulfoxonium reagents, highlighting their similarities and differences in terms of reaction speed and ease of preparation.", "Key Factual Information": "much slower and are typically easier to prepare", "Prefix reformulated": "How do stabilized ylides react compared to sulfoxonium reagents?", "Prefix-intent": "These , similarly to sulfoxonium reagents , react", "Verification": "The KFI directly describes the reaction speed and preparation ease of stabilized ylides, as prompted by the PI." }, { "Explanation": "The PI introduces a limitation of stabilized ylides, and the KFI provides specific details about this limitation and examples. The <|retrieval|> token is inserted to indicate where additional information about the reactivity of stabilized ylides might be needed.", "Ideal Knowledge": "Detailed information on the reactivity limitations of stabilized ylides, including specific examples with different functional groups (amides, esters, and other EWGs), and explanations of why certain reactions are more common or rare.", "Key Factual Information": "prohibitively sluggish : examples involving amides are widespread , with many fewer involving esters and virtually no examples involving other EWG 's", "Prefix reformulated": "What limitation do stabilized ylides have in their usefulness?", "Prefix-intent": "These are limited in their usefulness as the reaction can become", "Verification": "The KFI directly explains the limitation of stabilized ylides and provides examples, as prompted by the PI." }, { "Explanation": "The PI sets up the context for an alternative reaction, and the KFI provides the specific name of this reaction. The <|retrieval|> token is inserted to indicate where additional information about the Darzens reaction might be needed.", "Ideal Knowledge": "Information on the Darzens reaction, including its mechanism, applications, and advantages over stabilized ylide reactions, particularly for cases where ylide reactions are too slow or inefficient.", "Key Factual Information": "Darzens reaction is typically more appropriate", "Prefix reformulated": "What related reaction is more appropriate for stabilized ylides with limited reactivity?", "Prefix-intent": "For these , the related", "Verification": "The KFI directly provides an alternative reaction that is more suitable, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n If the ylide carbon is substituted with an aryl or allyl group , the reagent is referred to as a semi @-@ stabilized ylide . These have been developed extensively , second only to the classical methylene reagents ( R1 = R2 = H ) . The substitution pattern on aryl reagents can heavily influence the selectivity of the reaction as per the criteria above . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n If the ylide carbon is substituted with an alkyl group the reagent is referred to as an unstabilized ylide . The size of the alkyl groups are the major factors in selectivity with these reagents . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 R @-@ groups on the sulfur , though typically methyls , have been used to synthesize reagents that can perform enantioselective variants of the reaction ( See Variations below ) . The size of the groups can also influence diastereoselectivity in alicyclic substrates . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 R @-@ groups on the sulfur , though typically methyls , have been used to synthesize reagents that can perform <|end-PI|> <|retrieval|> <|start-KFI|> enantioselective variants of the reaction <|end-KFI|> ( See Variations below ) . <|start-PI|> The size of the groups can also influence <|end-PI|> <|retrieval|> <|start-KFI|> diastereoselectivity in alicyclic substrates <|end-KFI|> .", "none": null, "results": [ { "Explanation": "The PI introduces the R-groups on sulfur and sets up a query about their use in synthesis, while the KFI provides the specific type of reaction these reagents can perform. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.", "Ideal Knowledge": "A detailed explanation of how R-groups on sulfur, particularly methyl groups, can be modified to create reagents capable of performing enantioselective reactions, including examples of such reactions and their mechanisms.", "Key Factual Information": "enantioselective variants of the reaction", "Prefix reformulated": "What types of reactions can the R-groups on sulfur be used to synthesize reagents for?", "Prefix-intent": "The R @-@ groups on the sulfur , though typically methyls , have been used to synthesize reagents that can perform", "Verification": "The KFI directly provides information about the type of reactions that can be performed using reagents synthesized from R-groups on sulfur, as prompted by the PI." }, { "Explanation": "The PI sets up a query about the influence of group size, and the KFI provides the specific effect on diastereoselectivity in alicyclic substrates. The <|retrieval|> token is inserted to indicate where additional information about the relationship between group size and stereoselectivity might be needed.", "Ideal Knowledge": "A comprehensive explanation of how the size of R-groups on sulfur affects diastereoselectivity in reactions involving alicyclic substrates, including examples of different group sizes and their corresponding effects on stereochemistry.", "Key Factual Information": "diastereoselectivity in alicyclic substrates", "Prefix reformulated": "How does the size of R-groups on sulfur influence reactions?", "Prefix-intent": "The size of the groups can also influence", "Verification": "The KFI directly provides information about how the size of the groups influences a specific aspect of reactions with certain substrates, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information.\n\nHere is the passage you will be working with:\n\n<passage>\n Reactions of sulfur ylides with ketones and aldehydes to form epoxides are by far the most common application of the Johnson – Corey – Chaykovsky reaction . Examples involving complex substrates and ' exotic ' ylides have been reported , as shown below . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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|>Reactions of sulfur ylides with ketones and aldehydes to form epoxides are<|end-PI|> <|retrieval|> <|start-KFI|>by far the most common application of the Johnson – Corey – Chaykovsky reaction<|end-KFI|>. <|start-PI|>Examples involving complex substrates and ' exotic ' ylides have been<|end-PI|> <|retrieval|> <|start-KFI|>reported , as shown below<|end-KFI|>.", "none": null, "results": [ { "Explanation": "The PI introduces the topic of sulfur ylide reactions, and the KFI specifies that this is the most common application of the Johnson – Corey – Chaykovsky reaction. The <|retrieval|> token is inserted to indicate where additional information about the reaction's applications might be needed.", "Ideal Knowledge": "A comprehensive overview of the Johnson – Corey – Chaykovsky reaction, including its various applications and the prominence of sulfur ylide reactions with ketones and aldehydes to form epoxides.", "Key Factual Information": "by far the most common application of the Johnson – Corey – Chaykovsky reaction", "Prefix reformulated": "What is the most common application of the Johnson – Corey – Chaykovsky reaction?", "Prefix-intent": "Reactions of sulfur ylides with ketones and aldehydes to form epoxides are", "Verification": "The KFI directly provides information about the most common application of the reaction mentioned in the PI." }, { "Explanation": "The PI sets up an expectation for examples of complex reactions, and the KFI confirms that such examples have been reported. The <|retrieval|> token is inserted to indicate where additional information about these specific examples might be needed.", "Ideal Knowledge": "A collection of research papers or case studies demonstrating the use of complex substrates and 'exotic' ylides in the Johnson – Corey – Chaykovsky reaction, including specific examples and their outcomes.", "Key Factual Information": "reported , as shown below", "Prefix reformulated": "What has been reported about examples involving complex substrates and 'exotic' ylides in the Johnson – Corey – Chaykovsky reaction?", "Prefix-intent": "Examples involving complex substrates and ' exotic ' ylides have been", "Verification": "The KFI confirms that examples of complex substrates and 'exotic' ylides have been reported, as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 reaction has been used in a number of notable total syntheses including the Danishefsky Taxol total synthesis , which produces the chemotherapeutic drug taxol , and the Kuehne Strychnine total synthesis which produces the pesticide strychnine . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 synthesis of aziridines from imines is another important application of the Johnson – Corey – Chaykovsky reaction and provides an alternative to amine transfer from oxaziridines . Though less widely applied , the reaction has a similar substrate scope and functional group tolerance to the carbonyl equivalent . The examples shown below are representative ; in the latter , an aziridine forms in situ and is opened via nucleophilic attack to form the corresponding amine . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 synthesis of aziridines from imines is another important application of<|end-PI|> <|retrieval|> <|start-KFI|>the Johnson – Corey – Chaykovsky reaction and provides an alternative to amine transfer from oxaziridines<|end-KFI|>. <|start-PI|>Though less widely applied , the reaction has a similar<|end-PI|> <|retrieval|> <|start-KFI|>substrate scope and functional group tolerance to the carbonyl equivalent<|end-KFI|>. <|start-PI|>The examples shown below are representative ; in the latter , an aziridine forms in situ and<|end-PI|> <|retrieval|> <|start-KFI|>is opened via nucleophilic attack to form the corresponding amine<|end-KFI|>.", "none": null, "results": [ { "Explanation": "The PI introduces the topic of aziridine synthesis from imines, while the KFI provides the specific reaction name and its significance. The <|retrieval|> token is inserted to indicate where additional information about the reaction might be needed.", "Ideal Knowledge": "Detailed information about the Johnson – Corey – Chaykovsky reaction, its applications in aziridine synthesis, and comparisons with other methods like amine transfer from oxaziridines.", "Key Factual Information": "the Johnson – Corey – Chaykovsky reaction and provides an alternative to amine transfer from oxaziridines", "Prefix reformulated": "What is another important application of the synthesis of aziridines from imines?", "Prefix-intent": "The synthesis of aziridines from imines is another important application of", "Verification": "The KFI directly provides the specific reaction and alternative method related to the synthesis mentioned in the PI." }, { "Explanation": "The PI sets up a comparison of the reaction's characteristics, and the KFI provides specific details about its scope and tolerance. The <|retrieval|> token is inserted to indicate where additional information about reaction characteristics might be needed.", "Ideal Knowledge": "Comparative analysis of the Johnson – Corey – Chaykovsky reaction and carbonyl reactions, focusing on substrate scope and functional group tolerance.", "Key Factual Information": "substrate scope and functional group tolerance to the carbonyl equivalent", "Prefix reformulated": "What characteristics does the Johnson – Corey – Chaykovsky reaction have despite being less widely applied?", "Prefix-intent": "Though less widely applied , the reaction has a similar", "Verification": "The KFI directly describes the characteristics of the reaction as prompted by the PI." }, { "Explanation": "The PI introduces the formation of aziridine in situ, and the KFI provides information about its subsequent reaction. The <|retrieval|> token is inserted to indicate where additional information about aziridine reactions might be needed.", "Ideal Knowledge": "Detailed explanation of the in situ formation of aziridines and their subsequent reactions, specifically focusing on nucleophilic ring-opening to form amines.", "Key Factual Information": "is opened via nucleophilic attack to form the corresponding amine", "Prefix reformulated": "What happens to the aziridine that forms in situ in the latter example?", "Prefix-intent": "The examples shown below are representative ; in the latter , an aziridine forms in situ and", "Verification": "The KFI directly describes the process that happens to the aziridine as prompted by the PI." } ] }

[ { "content": "You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details.\n\nYour goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., \"The president of the USA is\"), but before the key factual information is stated (e.g., \"Joe Biden\"). Whenever there is 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 addition of sulfur ylides to enones , higher 1 @,@ 4 @-@ selectivity is typically obtained with sulfoxonium reagents than with sulfonium reagents . Many electron @-@ withdrawing groups have been shown compatible with the reaction including ketones , esters , and amides ( the example below involves a Weinreb amide ) . With further conjugated systems 1 @,@ 6 @-@ addition tends to predominate over 1 @,@ 4 @-@ addition . \n\n</passage>\n\nFollow these steps to complete the task:\n\n1. Inserting retrieval tokens:\n a. Read through the text carefully.\n b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words.\n c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI).\n d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI).\n e. Key Factual information should never intersect with Prefix Intent. Ensure this!\n\n2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI):\n a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI).\n b. Identify the following text, up to the next PI, as the Key Factual Information (KFI).\n c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words.\n\n3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>.\n a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>, <|end-PI|>) to quickly identify them.\n b. For each identified Key Factual Information (KFI), also add a start and end special tokens (<|start-KFI|>, <|end-KFI|>) \n c. Ensure that Prefix Intent (PI) and Key Factual Information (KFI) special tokens are inserted.\n\n4. Describing ideal retrieved information:\n a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. \n b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. \n c. Describe how that content should look like so it helps generating the Key Factual Information (KFI)\n\n\n5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent.\n\n\n6. Output format:\n For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON.\n\n - Prefix-intent: [Write the Prefix Intent here]\n - Prefix reformulated: [Write the Prefix Intent reformulated]\n - Key Factual Information: [Write the Key Factual Information here]\n - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context]\n - Ideal Knowledge: [Describe how the ideal retrieved content should look to generate the KFI]\n - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position]\n\n\n7. Final step:\n After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions and also <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|> special tokens. Ensure that PI and KFI special tokens are inside the updated passage.\n\nHere's an example of the expected output format. Updated Passage is constructed using PI-KFI pairs plus the additional special tokens:\n\nPassage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\nOutput:\n{ \"results\" : [\n {\n \"Prefix-intent\": \"Girmal Falls General This waterfall extends to a height of up to\",\n \"Prefix reformulated\": \"What is the height of Girmal Falls?\",\n \"Key Factual Information\": \"100 feet, making it the highest waterfall of Gujarat\",\n \"Verification\": \"The KFI directly provides the height information requested in the PI.\",\n \"Ideal Knowledge\": \"A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.\",\n \"Explanation\": \"The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement.\"\n },\n {\n \"Prefix-intent\": \"The water swiftly falls from a great height, creating\",\n \"Prefix reformulated\": \"What does the water falling from Girmal Falls create?\",\n \"Key Factual Information\": \"a fog like condition that's eye catching\",\n \"Verification\": \"The KFI directly describes the effect created by the falling water, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.\",\n \"Explanation\": \"The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description.\"\n },\n {\n \"Prefix-intent\": \"The fall comes to its best form at the time of\",\n \"Prefix reformulated\": \"When does Girmal Falls come to its best form?\",\n \"Key Factual Information\": \"monsoon and provides an immensely striking appearance\",\n \"Verification\": \"The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.\",\n \"Ideal Knowledge\": \"Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.\",\n \"Explanation\": \"The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed.\"\n }],\n\"Updated_Passage\":\"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler.\"\n}\n\n \nRemember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens 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 }