aorogat/question-to-tagged-question-llama3-lora
Token Classification
•
Updated
•
37
•
1
instruction
stringclasses 1
value | input
stringlengths 16
331
| output
stringlengths 30
443
|
|---|---|---|
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Is the organisation "Davis Brody Bond" the architect of the building "National September 11 Memorial & Museum"? | <qt>Is</qt> <s>the organisation "Davis Brody Bond"</s> <p>the architect of</p> <o>the building "National September 11 Memorial & Museum"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Is the organisation "1969 uprising in East Pakistan" the architect of the building "National September 11 Memorial & Museum"? | <qt>Is</qt> <s>the organisation "1969 uprising in East Pakistan"</s> <p>the architect of</p> <o>the building "National September 11 Memorial & Museum"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | What is the architect of the building "National September 11 Memorial & Museum"? | <qt>What</qt> <p>is the architect of</p> <o>the building "National September 11 Memorial & Museum"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Give me the architect of the building "National September 11 Memorial & Museum"? | <qt>Give me</qt> <p>the architect of</p> <o>the building "National September 11 Memorial & Museum"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | The architect of the building "National September 11 Memorial & Museum"? | <p>the architect of</p> <o>the building "National September 11 Memorial & Museum"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Which place has at least 2 architects? | <qt>Which</qt> <t>place</t> has <op>at least</op> <o>2</o> <p>architects</p>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Which place has at least 6 architects? | <qt>Which</qt> <t>place</t> has <op>at least</op> <o>6</o> <p>architects</p>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Is the organisation "Snøhetta" the architect of the building "National September 11 Memorial & Museum"? | <qt>Is</qt> <s>the organisation "Snøhetta"</s> <p>the architect of</p> <o>the building "National September 11 Memorial & Museum"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Is the organisation "1969 uprising in East Pakistan" the architect of the building "National September 11 Memorial & Museum"? | <qt>Is</qt> <s>the organisation "1969 uprising in East Pakistan"</s> <p>the architect of</p> <o>the building "National September 11 Memorial & Museum"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | What is the architect of the building "National September 11 Memorial & Museum"? | <qt>What</qt> <p>is the architect of</p> <o>the building "National September 11 Memorial & Museum"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Find the architect of the building "National September 11 Memorial & Museum"? | <qt>Find</qt> <p>the architect of</p> <o>the building "National September 11 Memorial & Museum"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | The architect of the building "National September 11 Memorial & Museum"? | <p>the architect of</p> <o>the building "National September 11 Memorial & Museum"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Which place has at least 2 architects? | <qt>Which</qt> <t>place</t> has <op>at least</op> <o>2</o> <p>architects</p>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Which place has at least 6 architects? | <qt>Which</qt> <t>place</t> has <op>at least</op> <o>6</o> <p>architects</p>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Is the organisation "HKS, Inc." the architect of the building "Guaranteed Rate Field"? | <qt>Is</qt> <s>the organisation "HKS, Inc."</s> <p>the architect of</p> <o>the building "Guaranteed Rate Field"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Is the organisation "1969 uprising in East Pakistan" the architect of the building "Guaranteed Rate Field"? | <qt>Is</qt> <s>the organisation "1969 uprising in East Pakistan"</s> <p>the architect of</p> <o>the building "Guaranteed Rate Field"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | What is the architect of the building "Guaranteed Rate Field"? | <qt>What</qt> <p>is the architect of</p> <o>the building "Guaranteed Rate Field"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Give me the architect of the building "Guaranteed Rate Field"? | <qt>Give me</qt> <p>the architect of</p> <o>the building "Guaranteed Rate Field"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | The architect of the building "Guaranteed Rate Field"? | <p>the architect of</p> <o>the building "Guaranteed Rate Field"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Which venues whose seating capacity is 40615 and whose tenant is Chicago White Sox? | <qt>Which</qt> <t>venues</t> whose <p>seating capacity</p> is <o>40615</o> <cc>and</cc> whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | What are the venues whose seating capacity is 40615 and whose tenant is Chicago White Sox? | <qt>What</qt> are the <t>venues</t> whose <p>seating capacity</p> is <o>40615</o> <cc>and</cc> whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Tell me venues whose seating capacity is 40615 and whose tenant is Chicago White Sox? | <qt>Tell me</qt> <t>venues</t> whose <p>seating capacity</p> is <o>40615</o> <cc>and</cc> whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Venues whose seating capacity is 40615 and whose tenant is Chicago White Sox? | <t>venues</t> whose <p>seating capacity</p> is <o>40615</o> <cc>and</cc> whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Which venues either whose seating capacity is 40615 or whose tenant is Chicago White Sox? | <qt>Which</qt> <t>venues</t> either whose <p>seating capacity</p> is <o>40615</o> <cc>or</cc> whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | What are the venues either whose seating capacity is 40615 or whose tenant is Chicago White Sox? | <qt>What</qt> are the <t>venues</t> either whose <p>seating capacity</p> is <o>40615</o> <cc>or</cc> whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Give me venues either whose seating capacity is 40615 or whose tenant is Chicago White Sox? | <qt>Give me</qt> <t>venues</t> either whose <p>seating capacity</p> is <o>40615</o> <cc>or</cc> whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Venues either whose seating capacity is 40615 or whose tenant is Chicago White Sox? | <t>venues</t> either whose <p>seating capacity</p> is <o>40615</o> <cc>or</cc> whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | How many venues whose seating capacity is 40615 and whose tenant is Chicago White Sox? | <qt>How many</qt> <t>venues</t> whose <p>seating capacity</p> is <o>40615</o> <cc>and</cc> whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | How many venues either whose seating capacity is 40615 or whose tenant is Chicago White Sox? | <qt>How many</qt> <t>venues</t> either whose <p>seating capacity</p> is <o>40615</o> <cc>or</cc> whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | How many venues whose seating capacity is 40615 or whose tenant is not Chicago White Sox? | <qt>How many</qt> <t>venues</t> whose <p>seating capacity</p> is <o>40615</o> or whose <p>tenant</p> is not <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | How many venues neither whose seating capacity is 40615 nor whose tenant is Chicago White Sox? | <qt>How many</qt> <t>venues</t> neither whose <p>seating capacity</p> is <o>40615</o> <cc>nor</cc> whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Which venues whose seating capacity is 40615, whose tenant is Chicago White Sox, and whose location is Chicago? | <qt>Which</qt> <t>venues</t> whose <p>seating capacity</p> is <o>40615</o>, whose <p>tenant</p> is <o>Chicago White Sox</o>, and whose <p>location</p> is <o>Chicago</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | What are the venues whose seating capacity is 40615, whose tenant is Chicago White Sox, and whose location is Chicago? | <qt>What</qt> are the <t>venues</t> whose <p>seating capacity</p> is <o>40615</o>, whose <p>tenant</p> is <o>Chicago White Sox</o>, and whose <p>location</p> is <o>Chicago</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Mention venues whose seating capacity is 40615, whose tenant is Chicago White Sox, and whose location is Chicago? | <qt>Mention</qt> <t>venues</t> whose <p>seating capacity</p> is <o>40615</o>, whose <p>tenant</p> is <o>Chicago White Sox</o>, and whose <p>location</p> is <o>Chicago</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Venues whose seating capacity is 40615, whose tenant is Chicago White Sox, and whose location is Chicago? | <t>venues</t> whose <p>seating capacity</p> is <o>40615</o>, whose <p>tenant</p> is <o>Chicago White Sox</o>, and whose <p>location</p> is <o>Chicago</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Which venues whose seating capacity is 40615, whose tenant is Chicago White Sox, or whose location is Chicago? | <qt>Which</qt> <t>venues</t> whose <p>seating capacity</p> is <o>40615</o>, whose <p>tenant</p> is <o>Chicago White Sox</o>, <cc>or</cc> whose <p>location</p> is <o>Chicago</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | What are the venues whose seating capacity is 40615, whose tenant is Chicago White Sox, or whose location is Chicago? | <qt>What</qt> are the <t>venues</t> whose <p>seating capacity</p> is <o>40615</o>, whose <p>tenant</p> is <o>Chicago White Sox</o>, <cc>or</cc> whose <p>location</p> is <o>Chicago</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Find venues whose seating capacity is 40615, whose tenant is Chicago White Sox, or whose location is Chicago? | <qt>Find</qt> <t>venues</t> whose <p>seating capacity</p> is <o>40615</o>, whose <p>tenant</p> is <o>Chicago White Sox</o>, <cc>or</cc> whose <p>location</p> is <o>Chicago</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Venues whose seating capacity is 40615, whose tenant is Chicago White Sox, or whose location is Chicago? | <t>venues</t> whose <p>seating capacity</p> is <o>40615</o>, whose <p>tenant</p> is <o>Chicago White Sox</o>, <cc>or</cc> whose <p>location</p> is <o>Chicago</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | How many venues whose seating capacity is 40615, whose tenant is Chicago White Sox, and whose location is Chicago? | <qt>How many</qt> <t>venues</t> whose <p>seating capacity</p> is <o>40615</o>, whose <p>tenant</p> is <o>Chicago White Sox</o>, and whose <p>location</p> is <o>Chicago</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | How many venues whose seating capacity is 40615, whose tenant is Chicago White Sox, or whose location is Chicago? | <qt>How many</qt> <t>venues</t> whose <p>seating capacity</p> is <o>40615</o>, whose <p>tenant</p> is <o>Chicago White Sox</o>, <cc>or</cc> whose <p>location</p> is <o>Chicago</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Which venues whose location is Chicago, whose architect is HKS, Inc., and whose tenant is Chicago White Sox? | <qt>Which</qt> <t>venues</t> whose <p>location</p> is <o>Chicago</o>, whose <p>architect</p> is <o>HKS, Inc.</o>, and whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | What are the venues whose location is Chicago, whose architect is HKS, Inc., and whose tenant is Chicago White Sox? | <qt>What</qt> are the <t>venues</t> whose <p>location</p> is <o>Chicago</o>, whose <p>architect</p> is <o>HKS, Inc.</o>, and whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Give me venues whose location is Chicago, whose architect is HKS, Inc., and whose tenant is Chicago White Sox? | <qt>Give me</qt> <t>venues</t> whose <p>location</p> is <o>Chicago</o>, whose <p>architect</p> is <o>HKS, Inc.</o>, and whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Venues whose location is Chicago, whose architect is HKS, Inc., and whose tenant is Chicago White Sox? | <t>venues</t> whose <p>location</p> is <o>Chicago</o>, whose <p>architect</p> is <o>HKS, Inc.</o>, and whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Which venues whose location is Chicago, whose architect is HKS, Inc., or whose tenant is Chicago White Sox? | <qt>Which</qt> <t>venues</t> whose <p>location</p> is <o>Chicago</o>, whose <p>architect</p> is <o>HKS, Inc.</o>, <cc>or</cc> whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | What are the venues whose location is Chicago, whose architect is HKS, Inc., or whose tenant is Chicago White Sox? | <qt>What</qt> are the <t>venues</t> whose <p>location</p> is <o>Chicago</o>, whose <p>architect</p> is <o>HKS, Inc.</o>, <cc>or</cc> whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Find venues whose location is Chicago, whose architect is HKS, Inc., or whose tenant is Chicago White Sox? | <qt>Find</qt> <t>venues</t> whose <p>location</p> is <o>Chicago</o>, whose <p>architect</p> is <o>HKS, Inc.</o>, <cc>or</cc> whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Venues whose location is Chicago, whose architect is HKS, Inc., or whose tenant is Chicago White Sox? | <t>venues</t> whose <p>location</p> is <o>Chicago</o>, whose <p>architect</p> is <o>HKS, Inc.</o>, <cc>or</cc> whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | How many venues whose location is Chicago, whose architect is HKS, Inc., and whose tenant is Chicago White Sox? | <qt>How many</qt> <t>venues</t> whose <p>location</p> is <o>Chicago</o>, whose <p>architect</p> is <o>HKS, Inc.</o>, and whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | How many venues whose location is Chicago, whose architect is HKS, Inc., or whose tenant is Chicago White Sox? | <qt>How many</qt> <t>venues</t> whose <p>location</p> is <o>Chicago</o>, whose <p>architect</p> is <o>HKS, Inc.</o>, <cc>or</cc> whose <p>tenant</p> is <o>Chicago White Sox</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Which venue has the most seating capacity after the venue "Coors Field"? | <qt>Which</qt> <t>venue</t> <off>has the most</off> <p>seating capacity</p> <op>after</op> <o>the venue "Coors Field"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | A venue which has the most seating capacity after the venue "Coors Field"? | <t>venue</t> <off>which has the most</off> <p>seating capacity</p> <op>after</op> <o>the venue "Coors Field"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Give me a venue which has the most seating capacity after the venue "Coors Field"? | <qt>Give me</qt> <t>venue</t> <off>which has the most</off> <p>seating capacity</p> <op>after</op> <o>the venue "Coors Field"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Which venue after the venue "Coors Field" has the second most seating capacity? | <qt>Which</qt> <t>venue</t> <op>after</op> <o>the venue "Coors Field"</o> <off>has the second most</off> <p>seating capacity</p>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | A venue after the venue "Coors Field" which has the second most seating capacity? | <t>venue</t> <op>after</op> <o>the venue "Coors Field"</o> <off>which has the second most</off> <p>seating capacity</p>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Return a venue after the venue "Coors Field" which has the second most seating capacity? | <qt>Return</qt> <t>venue</t> <op>after</op> <o>the venue "Coors Field"</o> <off>which has the second most</off> <p>seating capacity</p>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Who is the founder of the location of Guaranteed Rate Field? | <qt>Who</qt> <p>is the founder of</p> <p>the location of</p> <o>Guaranteed Rate Field</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Tell me the founder of the location of Guaranteed Rate Field? | <qt>Tell me</qt> <p>the founder of</p> <p>the location of</p> <o>Guaranteed Rate Field</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | The founder of the location of Guaranteed Rate Field? | <p>the founder of</p> <p>the location of</p> <o>Guaranteed Rate Field</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | What is the occupation of the founder of the location of Guaranteed Rate Field? | <qt>What</qt> <p>is the occupation of</p> <p>the founder of</p> <p>the location of</p> <o>Guaranteed Rate Field</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Tell me the occupation of the founder of the location of Guaranteed Rate Field? | <qt>Tell me</qt> <p> the occupation of</p> <p>the founder of</p> <p>the location of</p> <o>Guaranteed Rate Field</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | The occupation of the founder of the location of Guaranteed Rate Field? | <p>the occupation of</p> <p>the founder of</p> <p>the location of</p> <o>Guaranteed Rate Field</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Is the organisation "New South Wales Government Architect" the architect of the building "Old Government House, Parramatta"? | <qt>Is</qt> <s>the organisation "New South Wales Government Architect"</s> <p>the architect of</p> <o>the building "Old Government House, Parramatta"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Is the organisation "1969 uprising in East Pakistan" the architect of the building "Old Government House, Parramatta"? | <qt>Is</qt> <s>the organisation "1969 uprising in East Pakistan"</s> <p>the architect of</p> <o>the building "Old Government House, Parramatta"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | What is the architect of the building "Old Government House, Parramatta"? | <qt>What</qt> <p>is the architect of</p> <o>the building "Old Government House, Parramatta"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Mention the architect of the building "Old Government House, Parramatta"? | <qt>Mention</qt> <p>the architect of</p> <o>the building "Old Government House, Parramatta"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | The architect of the building "Old Government House, Parramatta"? | <p>the architect of</p> <o>the building "Old Government House, Parramatta"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Which buildings whose location is Sydney and whose architect is the "New South Wales Government Architect"? | <qt>Which</qt> <t>buildings</t> whose <p>location</p> is <o>Sydney</o> <cc>and</cc> whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | What are the buildings whose location is Sydney and whose architect is the "New South Wales Government Architect"? | <qt>What</qt> are the <t>buildings</t> whose <p>location</p> is <o>Sydney</o> <cc>and</cc> whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Give me buildings whose location is Sydney and whose architect is the "New South Wales Government Architect"? | <qt>Give me</qt> <t>buildings</t> whose <p>location</p> is <o>Sydney</o> <cc>and</cc> whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Buildings whose location is Sydney and whose architect is the "New South Wales Government Architect"? | <t>buildings</t> whose <p>location</p> is <o>Sydney</o> <cc>and</cc> whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Which buildings either whose location is Sydney or whose architect is the "New South Wales Government Architect"? | <qt>Which</qt> <t>buildings</t> either whose <p>location</p> is <o>Sydney</o> <cc>or</cc> whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | What are the buildings either whose location is Sydney or whose architect is the "New South Wales Government Architect"? | <qt>What</qt> are the <t>buildings</t> either whose <p>location</p> is <o>Sydney</o> <cc>or</cc> whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Find buildings either whose location is Sydney or whose architect is the "New South Wales Government Architect"? | <qt>Find</qt> <t>buildings</t> either whose <p>location</p> is <o>Sydney</o> <cc>or</cc> whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Buildings either whose location is Sydney or whose architect is the "New South Wales Government Architect"? | <t>buildings</t> either whose <p>location</p> is <o>Sydney</o> <cc>or</cc> whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Which buildings whose location is Sydney but whose architect is not the "New South Wales Government Architect"? | <qt>Which</qt> <t>buildings</t> whose <p>location</p> is <o>Sydney</o> <cc>but</cc> whose <p>architect</p> is not <o>the "New South Wales Government Architect"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | What are the buildings whose location is Sydney but whose architect is not the "New South Wales Government Architect"? | <qt>What</qt> are the <t>buildings</t> whose <p>location</p> is <o>Sydney</o> <cc>but</cc> whose <p>architect</p> is not <o>the "New South Wales Government Architect"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Mention buildings whose location is Sydney but whose architect is not the "New South Wales Government Architect"? | <qt>Mention</qt> <t>buildings</t> whose <p>location</p> is <o>Sydney</o> <cc>but</cc> whose <p>architect</p> is not <o>the "New South Wales Government Architect"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Buildings whose location is Sydney but whose architect is not the "New South Wales Government Architect"? | <t>buildings</t> whose <p>location</p> is <o>Sydney</o> <cc>but</cc> whose <p>architect</p> is not <o>the "New South Wales Government Architect"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | How many buildings whose location is Sydney and whose architect is the "New South Wales Government Architect"? | <qt>How many</qt> <t>buildings</t> whose <p>location</p> is <o>Sydney</o> <cc>and</cc> whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | How many buildings either whose location is Sydney or whose architect is the "New South Wales Government Architect"? | <qt>How many</qt> <t>buildings</t> either whose <p>location</p> is <o>Sydney</o> <cc>or</cc> whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | How many buildings whose location is Sydney but whose architect is not the "New South Wales Government Architect"? | <qt>How many</qt> <t>buildings</t> whose <p>location</p> is <o>Sydney</o> <cc>but</cc> whose <p>architect</p> is not <o>the "New South Wales Government Architect"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | How many buildings whose location is Sydney or whose architect is not the "New South Wales Government Architect"? | <qt>How many</qt> <t>buildings</t> whose <p>location</p> is <o>Sydney</o> or whose <p>architect</p> is not <o>the "New South Wales Government Architect"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | How many buildings neither whose location is Sydney nor whose architect is the "New South Wales Government Architect"? | <qt>How many</qt> <t>buildings</t> neither whose <p>location</p> is <o>Sydney</o> <cc>nor</cc> whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Which building has at least 6 locations? | <qt>Which</qt> <t>building</t> has <op>at least</op> <o>6</o> <p>locations</p>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Which buildings whose location is Sydney, whose architect is the "New South Wales Government Architect", and whose owner is the "National Trust of Australia"? | <qt>Which</qt> <t>buildings</t> whose <p>location</p> is <o>Sydney</o>, whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>, and whose <p>owner</p> is <o>the "National Trust of Australia"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | What are the buildings whose location is Sydney, whose architect is the "New South Wales Government Architect", and whose owner is the "National Trust of Australia"? | <qt>What</qt> are the <t>buildings</t> whose <p>location</p> is <o>Sydney</o>, whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>, and whose <p>owner</p> is <o>the "National Trust of Australia"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Mention buildings whose location is Sydney, whose architect is the "New South Wales Government Architect", and whose owner is the "National Trust of Australia"? | <qt>Mention</qt> <t>buildings</t> whose <p>location</p> is <o>Sydney</o>, whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>, and whose <p>owner</p> is <o>the "National Trust of Australia"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Buildings whose location is Sydney, whose architect is the "New South Wales Government Architect", and whose owner is the "National Trust of Australia"? | <t>buildings</t> whose <p>location</p> is <o>Sydney</o>, whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>, and whose <p>owner</p> is <o>the "National Trust of Australia"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Which buildings whose location is Sydney, whose architect is the "New South Wales Government Architect", or whose owner is the "National Trust of Australia"? | <qt>Which</qt> <t>buildings</t> whose <p>location</p> is <o>Sydney</o>, whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>, <cc>or</cc> whose <p>owner</p> is <o>the "National Trust of Australia"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | What are the buildings whose location is Sydney, whose architect is the "New South Wales Government Architect", or whose owner is the "National Trust of Australia"? | <qt>What</qt> are the <t>buildings</t> whose <p>location</p> is <o>Sydney</o>, whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>, <cc>or</cc> whose <p>owner</p> is <o>the "National Trust of Australia"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Tell me buildings whose location is Sydney, whose architect is the "New South Wales Government Architect", or whose owner is the "National Trust of Australia"? | <qt>Tell me</qt> <t>buildings</t> whose <p>location</p> is <o>Sydney</o>, whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>, <cc>or</cc> whose <p>owner</p> is <o>the "National Trust of Australia"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Buildings whose location is Sydney, whose architect is the "New South Wales Government Architect", or whose owner is the "National Trust of Australia"? | <t>buildings</t> whose <p>location</p> is <o>Sydney</o>, whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>, <cc>or</cc> whose <p>owner</p> is <o>the "National Trust of Australia"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | How many buildings whose location is Sydney, whose architect is the "New South Wales Government Architect", and whose owner is the "National Trust of Australia"? | <qt>How many</qt> <t>buildings</t> whose <p>location</p> is <o>Sydney</o>, whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>, and whose <p>owner</p> is <o>the "National Trust of Australia"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | How many buildings whose location is Sydney, whose architect is the "New South Wales Government Architect", or whose owner is the "National Trust of Australia"? | <qt>How many</qt> <t>buildings</t> whose <p>location</p> is <o>Sydney</o>, whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>, <cc>or</cc> whose <p>owner</p> is <o>the "National Trust of Australia"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Which buildings whose location is New South Wales, Parramatta and Sydney and whose architect is the "New South Wales Government Architect"? | <qt>Which</qt> <t>buildings</t> whose <p>location</p> is <o>New South Wales</o>, <o>Parramatta</o> and <o>Sydney</o> <cc>and</cc> whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | What are the buildings whose location is New South Wales, Parramatta and Sydney and whose architect is the "New South Wales Government Architect"? | <qt>What</qt> are the <t>buildings</t> whose <p>location</p> is <o>New South Wales</o>, <o>Parramatta</o> and <o>Sydney</o> <cc>and</cc> whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Find buildings whose location is New South Wales, Parramatta and Sydney and whose architect is the "New South Wales Government Architect"? | <qt>Find</qt> <t>buildings</t> whose <p>location</p> is <o>New South Wales</o>, <o>Parramatta</o> and <o>Sydney</o> <cc>and</cc> whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>? |
### Instruction:
You are a powerful model trained to convert questions to tagged questions. Use the tags as follows:
<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests.
<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase.
<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before.
<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc.
<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate.
<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?'
<t> to surround entity types like person, place, etc.
<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc.
<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>). | Buildings whose location is New South Wales, Parramatta and Sydney and whose architect is the "New South Wales Government Architect"? | <t>buildings</t> whose <p>location</p> is <o>New South Wales</o>, <o>Parramatta</o> and <o>Sydney</o> <cc>and</cc> whose <p>architect</p> is <o>the "New South Wales Government Architect"</o>? |
End of preview. Expand
in Dataset Viewer.
Questions_to_Tagged_Questions_Prompts
Overview
The Questions_to_Tagged_Questions_Prompts dataset is designed to facilitate the training of the aorogat/question-to-tagged-question-llama3-lora model. This dataset consists of prompt files derived from the QueryBridge dataset, which includes one million annotated questions with SPARQL queries. The primary goal is to enhance the capabilities of question-answering (QA) systems over knowledge graphs by converting natural language questions into tagged question formats.
Dataset Source
The dataset is based on QueryBridge: One Million Annotated Questions with SPARQL Queries. This comprehensive resource enables the development and evaluation of algorithms that can process and interpret natural language questions within structured knowledge graphs.
Dataset Structure
The Questions_to_Tagged_Questions_Prompts dataset is organized as a JSON array, where each entry represents a prompt for training the model. Each prompt consists of:
- Instruction: Guidelines on how to convert questions into tagged questions.
- Input: A natural language question.
- Output: The corresponding tagged question format.
Example of JSON Object
{
"instruction": "### Instruction: \nYou are a powerful model trained to convert questions to tagged questions. Use the tags as follows: \n<qt> to surround question keywords like 'What', 'Who', 'Which', 'How many', 'Return' or any word that represents requests. \n<o> to surround entities as an object like person name, place name, etc. It must be a noun or a noun phrase. \n<s> to surround entities as a subject like person name, place name, etc. The difference between <s> and <o>, <s> only appear in yes/no questions as in the training data you saw before. \n<cc> to surround coordinating conjunctions that connect two or more phrases like 'and', 'or', 'nor', etc. \n<p> to surround predicates that may be an entity attribute or a relationship between two entities. It can be a verb phrase or a noun phrase. The question must contain at least one predicate. \n<off> for offset in questions asking for the second, third, etc. For example, the question 'What is the second largest country?', <off> will be located as follows. 'What is the <off>second</off> largest country?' \n<t> to surround entity types like person, place, etc. \n<op> to surround operators that compare quantities or values, like 'greater than', 'more than', etc. \n<ref> to indicate a reference within the question that requires a cycle to refer back to an entity (e.g., 'Who is the CEO of a company founded by himself?' where 'himself' would be tagged as <ref>himself</ref>).",
"input": "Is the organisation \"Davis Brody Bond\" the architect of the building \"National September 11 Memorial & Museum\"?",
"output": "<qt>Is</qt> <s>the organisation \"Davis Brody Bond\"</s> <p>the architect of</p> <o>the building \"National September 11 Memorial & Museum\"</o>?"
}
- Downloads last month
- 34
Size of downloaded dataset files:
1.76 GB
Size of the auto-converted Parquet files:
42.3 MB
Number of rows:
1,004,534
Models trained or fine-tuned on aorogat/Questions_to_Tagged_Questions_Prompts