id
stringlengths 1
12
| question
stringlengths 22
566
| golden_answers
dict | data_source
stringclasses 2
values | prompt
listlengths 1
1
| ability
stringclasses 6
values | reward_model
dict | extra_info
dict | metadata
dict |
|---|---|---|---|---|---|---|---|---|
35951
|
Aeronáutica provides consumer services for airlines.
|
{
"evidence": [
"Aeronautica_sentence_0"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nAeronáutica provides consumer services for airlines.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Aeronautica_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3884,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.701",
"value": "?search="
},
{
"operation": "search",
"time": "14.263",
"value": "Aeronautica"
},
{
"operation": "Now on",
"time": "15.149",
"value": "Aeronautica"
},
{
"operation": "hyperlink",
"time": "26.137",
"value": "Aeronáutica (Angola)"
},
{
"operation": "Now on",
"time": "26.752",
"value": "Aeronáutica (Angola)"
},
{
"operation": "search",
"time": "45.384",
"value": "Aeronáutica"
},
{
"operation": "Now on",
"time": "46.151",
"value": "Aeronautica"
},
{
"operation": "hyperlink",
"time": "53.55",
"value": "Aeronáutica Industrial S.A."
},
{
"operation": "Now on",
"time": "54.143",
"value": "Aeronáutica Industrial S.A."
},
{
"operation": "search",
"time": "69.942",
"value": "Aeronáutica provides consumer services for airlines"
},
{
"operation": "Now on",
"time": "71.408",
"value": "undefined"
},
{
"operation": "search",
"time": "81.875",
"value": "consumer service for airlines"
},
{
"operation": "Now on",
"time": "83.191",
"value": "undefined"
},
{
"operation": "search",
"time": "101.166",
"value": "Aeronautica"
},
{
"operation": "Now on",
"time": "102.15",
"value": "Aeronautica"
},
{
"operation": "hyperlink",
"time": "110.198",
"value": "Aeronáutica (Seville Metro)"
},
{
"operation": "Now on",
"time": "110.793",
"value": "Aeronáutica (Seville Metro)"
},
{
"operation": "back-button-clicked",
"time": "116.381",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "116.466",
"value": "Aeronautica"
},
{
"operation": "Highlighting",
"time": "118.692",
"value": "Aeronautica_sentence_0"
},
{
"operation": "finish",
"time": "129.173",
"value": "finish"
}
]
}
|
exfever_903
|
The Formula was the second mystery film directed by John G. Avildsen.
|
{
"evidence": [
"The Formula (1980 film)",
"John Gielgud"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Formula was the second mystery film directed by John G. Avildsen.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Formula (1980 film)",
"John Gielgud"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3885,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_663
|
Champion is a 1949 American drama film starring an American actress and entertainer born on August 3, 1921.
|
{
"evidence": [
"Champion (1949 film)",
"Marilyn Maxwell"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nChampion is a 1949 American drama film starring an American actress and entertainer born on August 3, 1921.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Champion (1949 film)",
"Marilyn Maxwell"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3886,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
79679
|
Dorsami Naidu had a misunderstanding with Pramod Rae for asking him publicly to resign when he has already given him private assurance that he would do so.
|
{
"evidence": [
"Dorsami Naidu_sentence_5",
"Dorsami Naidu_sentence_6"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nDorsami Naidu had a misunderstanding with Pramod Rae for asking him publicly to resign when he has already given him private assurance that he would do so.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Dorsami Naidu_sentence_5",
"Dorsami Naidu_sentence_6"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3887,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.011",
"value": "?search="
},
{
"operation": "search",
"time": "8.778",
"value": "Dorsami Naidu"
},
{
"operation": "Now on",
"time": "9.877",
"value": "Dorsami Naidu"
},
{
"operation": "Page search",
"time": "15.951",
"value": "mkisund"
},
{
"operation": "Page search",
"time": "19.131",
"value": "misund"
},
{
"operation": "search",
"time": "27.295",
"value": "Pramod Rae"
},
{
"operation": "Now on",
"time": "28.101",
"value": "Pramod Rae"
},
{
"operation": "Page search",
"time": "33.742",
"value": "dors"
},
{
"operation": "search",
"time": "39.684",
"value": "Dorsami Naidu had a misunderstanding with Pramod Rae"
},
{
"operation": "Now on",
"time": "41.109",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "42.769",
"value": "dorsami naidu had a misunderstanding with prasad rae"
},
{
"operation": "Now on",
"time": "43.414",
"value": "undefined"
},
{
"operation": "search",
"time": "48.534",
"value": "Dorsami Naidu had a misunderstanding"
},
{
"operation": "Now on",
"time": "49.235",
"value": "undefined"
},
{
"operation": "search",
"time": "55.314",
"value": "publicly to resign when he has already given him private assurance that he would do so"
},
{
"operation": "Now on",
"time": "60.011",
"value": "undefined"
},
{
"operation": "search",
"time": "64.74",
"value": "Dorsami Naidu"
},
{
"operation": "Now on",
"time": "65.571",
"value": "Dorsami Naidu"
},
{
"operation": "Page search",
"time": "68.766",
"value": "misund"
},
{
"operation": "Highlighting",
"time": "75.594",
"value": "Dorsami Naidu_sentence_5"
},
{
"operation": "Highlighting",
"time": "94.379",
"value": "Dorsami Naidu_sentence_6"
},
{
"operation": "finish",
"time": "99.612",
"value": "finish"
}
]
}
|
exfever_238
|
James Garner played the leadimg role in Great Escape with an American actor.
|
{
"evidence": [
"James Garner",
"Steve McQueen"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJames Garner played the leadimg role in Great Escape with an American actor.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"James Garner",
"Steve McQueen"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3888,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8028
|
"Just the Way You Are" is the debut solo single by Asian singer-songwriter Bruno Mars, who won a Grammy Award recognizing superior vocal performance by a male in the pop category, that also won by an American singer and musician, who is credited as a pioneer and influence by musicians across a range of genres that includes rhythm and blues.
|
{
"evidence": [
"Just the Way You Are (Bruno Mars song)",
"Grammy Award for Best Male Pop Vocal Performance",
"Stevie Wonder"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\n\"Just the Way You Are\" is the debut solo single by Asian singer-songwriter Bruno Mars, who won a Grammy Award recognizing superior vocal performance by a male in the pop category, that also won by an American singer and musician, who is credited as a pioneer and influence by musicians across a range of genres that includes rhythm and blues.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Just the Way You Are (Bruno Mars song)",
"Grammy Award for Best Male Pop Vocal Performance",
"Stevie Wonder"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3889,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
27859
|
Lodewijk van Bylandt(1718 – Hoeven, 28 December 1793) was a Dutch lieutenant-admiral who first served as a gunner then a captain of West Stellingwerff in 1756 and acting-commander of a Dutch naval expedition against Morocco.
|
{
"evidence": [
"Lodewijk van Bylandt_sentence_0",
"Lodewijk van Bylandt_sentence_10",
"Lodewijk van Bylandt_sentence_11"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nLodewijk van Bylandt(1718 – Hoeven, 28 December 1793) was a Dutch lieutenant-admiral who first served as a gunner then a captain of West Stellingwerff in 1756 and acting-commander of a Dutch naval expedition against Morocco.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Lodewijk van Bylandt_sentence_0",
"Lodewijk van Bylandt_sentence_10",
"Lodewijk van Bylandt_sentence_11"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3890,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.762",
"value": "undefined"
},
{
"operation": "search",
"time": "11.541",
"value": "Lodewijk van Bylandt(1718 – Hoeven, 28 December 1793)"
},
{
"operation": "Now on",
"time": "13.643",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "16.706",
"value": "Lodewijk van Bylandt"
},
{
"operation": "Now on",
"time": "17.281",
"value": "Lodewijk van Bylandt"
},
{
"operation": "Highlighting",
"time": "20.134",
"value": "Lodewijk van Bylandt_sentence_0"
},
{
"operation": "Page search",
"time": "33.076",
"value": "gunner"
},
{
"operation": "Page search",
"time": "58.95",
"value": "west"
},
{
"operation": "Highlighting",
"time": "67.767",
"value": "Lodewijk van Bylandt_sentence_10"
},
{
"operation": "Page search",
"time": "153.48",
"value": "morocco"
},
{
"operation": "Highlighting",
"time": "160.829",
"value": "Lodewijk van Bylandt_sentence_11"
},
{
"operation": "finish",
"time": "179.526",
"value": "finish"
}
]
}
|
50612
|
The most common type of split pin, also known as the cotter pin, is the extended prong with a diamond cut and hammer lock split pins are properly installed by striking the head with a hammer to secure the pin.
|
{
"evidence": [
"Split pin_sentence_12"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe most common type of split pin, also known as the cotter pin, is the extended prong with a diamond cut and hammer lock split pins are properly installed by striking the head with a hammer to secure the pin.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Split pin_sentence_12"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3891,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.142",
"value": "?search="
},
{
"operation": "search",
"time": "30.419",
"value": "Split pin"
},
{
"operation": "Now on",
"time": "31.397",
"value": "Split pin"
},
{
"operation": "Page search",
"time": "36.715",
"value": "cotter"
},
{
"operation": "Highlighting",
"time": "45.706",
"value": "Split pin_sentence_1"
},
{
"operation": "Page search",
"time": "55.388",
"value": "prong"
},
{
"operation": "Highlighting",
"time": "69.711",
"value": "Split pin_sentence_12"
},
{
"operation": "Highlighting deleted",
"time": "91.612",
"value": "Split pin_sentence_1"
},
{
"operation": "Page search",
"time": "97.041",
"value": "diamond"
},
{
"operation": "finish",
"time": "272.936",
"value": "finish"
}
]
}
|
74966
|
One Direction is an English-Irish boy band formed in 2010 who won the NME Award for Worst Band and whose single, "What Makes You Beautiful" ranked 1st in the UK and Irish chart in 2011
|
{
"evidence": [
"One Direction_sentence_1",
"What Makes You Beautiful_sentence_68",
"What Makes You Beautiful_sentence_72"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nOne Direction is an English-Irish boy band formed in 2010 who won the NME Award for Worst Band and whose single, \"What Makes You Beautiful\" ranked 1st in the UK and Irish chart in 2011\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"One Direction_sentence_1",
"What Makes You Beautiful_sentence_68",
"What Makes You Beautiful_sentence_72"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3892,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.966",
"value": "?search="
},
{
"operation": "search",
"time": "16.162",
"value": "One Direction"
},
{
"operation": "Now on",
"time": "18.318",
"value": "One Direction"
},
{
"operation": "Highlighting",
"time": "23.069",
"value": "One Direction_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "43.819",
"value": "One Direction_sentence_1"
},
{
"operation": "Page search",
"time": "47.877",
"value": "nme"
},
{
"operation": "Page search",
"time": "122.826",
"value": "what makes you"
},
{
"operation": "Page search",
"time": "224.435",
"value": "new musical"
},
{
"operation": "Highlighting",
"time": "280.657",
"value": "One Direction_sentence_1"
},
{
"operation": "Highlighting",
"time": "288.384",
"value": "One Direction_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "312.639",
"value": "One Direction_sentence_4"
},
{
"operation": "Page search",
"time": "334.585",
"value": "what makes"
},
{
"operation": "search",
"time": "370.914",
"value": "What Makes You Beautiful"
},
{
"operation": "Now on",
"time": "371.956",
"value": "What Makes You Beautiful"
},
{
"operation": "Highlighting",
"time": "398.221",
"value": "What Makes You Beautiful_sentence_68"
},
{
"operation": "Highlighting",
"time": "406.192",
"value": "What Makes You Beautiful_sentence_72"
},
{
"operation": "search",
"time": "430.042",
"value": "NME Awards one direction"
},
{
"operation": "Now on",
"time": "432.45",
"value": "undefined"
},
{
"operation": "search",
"time": "451.4",
"value": "NME Awards"
},
{
"operation": "Now on",
"time": "452.796",
"value": "NME Awards"
},
{
"operation": "search",
"time": "500.575",
"value": "nme award 2013"
},
{
"operation": "Now on",
"time": "501.73",
"value": "undefined"
},
{
"operation": "search",
"time": "515.663",
"value": "one direction nme"
},
{
"operation": "Now on",
"time": "516.21",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "528.055",
"value": "Massive Attack"
},
{
"operation": "Now on",
"time": "529.191",
"value": "Massive Attack"
},
{
"operation": "finish",
"time": "546.19",
"value": "finish"
}
]
}
|
exfever_443
|
Sierra Leone is infamous for Blood Diamonds.
|
{
"evidence": [
"Sierra Leone",
"West Africa"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nSierra Leone is infamous for Blood Diamonds.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Sierra Leone",
"West Africa"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3893,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
77893
|
Golden Resources Shopping Mall is a 557,419 square metres (6,000,010 sq ft) hotel completed on 20 October 2004.
|
{
"evidence": [
"Golden Resources Mall_sentence_2"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nGolden Resources Shopping Mall is a 557,419 square metres (6,000,010 sq ft) hotel completed on 20 October 2004.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Golden Resources Mall_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3894,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.307",
"value": "?search="
},
{
"operation": "Now on",
"time": "47.178",
"value": "?search="
},
{
"operation": "search",
"time": "192.87",
"value": "Golden Resources Mall"
},
{
"operation": "Now on",
"time": "193.973",
"value": "Golden Resources Mall"
},
{
"operation": "Highlighting",
"time": "213.425",
"value": "Golden Resources Mall_table_caption_0"
},
{
"operation": "Highlighting",
"time": "215.784",
"value": "Golden Resources Mall_header_cell_0_3_0"
},
{
"operation": "Highlighting",
"time": "216.903",
"value": "Golden Resources Mall_cell_0_3_1"
},
{
"operation": "Highlighting",
"time": "216.911",
"value": "Golden Resources Mall_header_cell_0_3_0"
},
{
"operation": "Highlighting",
"time": "324.943",
"value": "Golden Resources Mall_sentence_8"
},
{
"operation": "Highlighting",
"time": "402.551",
"value": "Golden Resources Mall_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "493.383",
"value": "Golden Resources Mall_cell_0_3_1"
},
{
"operation": "Highlighting deleted",
"time": "493.386",
"value": "Golden Resources Mall_header_cell_0_3_0"
},
{
"operation": "Highlighting deleted",
"time": "498.79",
"value": "Golden Resources Mall_table_caption_0"
},
{
"operation": "Highlighting",
"time": "501.334",
"value": "Golden Resources Mall_table_caption_0"
},
{
"operation": "Highlighting deleted",
"time": "502.47",
"value": "Golden Resources Mall_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "504.598",
"value": "Golden Resources Mall_table_caption_0"
},
{
"operation": "Highlighting deleted",
"time": "510.87",
"value": "Golden Resources Mall_sentence_8"
},
{
"operation": "hyperlink",
"time": "562.766",
"value": "shopping mall"
},
{
"operation": "Now on",
"time": "564.495",
"value": "Shopping mall"
},
{
"operation": "back-button-clicked",
"time": "603.104",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "603.334",
"value": "Golden Resources Mall"
},
{
"operation": "Highlighting",
"time": "627.87",
"value": "Golden Resources Mall_sentence_2"
},
{
"operation": "finish",
"time": "637.702",
"value": "finish"
}
]
}
|
7757
|
The 12002 Suess' discovery date was on 19 March 2000, and its discovery site was in Prague Obs. (the principal observatory of the Astronomical Institute).
|
{
"evidence": [
"12002 Suess_sentence_1"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe 12002 Suess' discovery date was on 19 March 2000, and its discovery site was in Prague Obs. (the principal observatory of the Astronomical Institute).\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"12002 Suess_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3895,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.522",
"value": "?search="
},
{
"operation": "search",
"time": "17.496",
"value": "12002 Suess"
},
{
"operation": "Now on",
"time": "18.42",
"value": "12002 Suess"
},
{
"operation": "Highlighting",
"time": "29.672",
"value": "12002 Suess_sentence_1"
},
{
"operation": "finish",
"time": "61.332",
"value": "finish"
}
]
}
|
exfever_7304
|
Virginia, in the Mid-Atlantic and Southeastern regions of the United States, is a constituent political entity, of which there are currently 60.
|
{
"evidence": [
"Virginia",
"U.S. state"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nVirginia, in the Mid-Atlantic and Southeastern regions of the United States, is a constituent political entity, of which there are currently 60.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Virginia",
"U.S. state"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3896,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5636
|
The Anschluss was the annexation of Austria into Nazi Germany began after the unification of Germany excluded Austria and the German Austrians from the Prussian-dominated period of the German Reich founded on 18 January 1875 also known as the Deutsche Reichsgründung.
|
{
"evidence": [
"Anschluss",
"German Empire",
"Proclamation of the German Empire"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Anschluss was the annexation of Austria into Nazi Germany began after the unification of Germany excluded Austria and the German Austrians from the Prussian-dominated period of the German Reich founded on 18 January 1875 also known as the Deutsche Reichsgründung.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Anschluss",
"German Empire",
"Proclamation of the German Empire"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3897,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6682
|
Saskatchewan is located in the middle of the three prairie sub-national administrative divisions within the geographical areas of the second-largest country in North America.
|
{
"evidence": [
"Saskatchewan",
"Provinces and territories of Canada",
"Canada"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nSaskatchewan is located in the middle of the three prairie sub-national administrative divisions within the geographical areas of the second-largest country in North America.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Saskatchewan",
"Provinces and territories of Canada",
"Canada"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3898,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4511
|
Tessanne Amanda Chin is a Canadian recording artist, best known for winning Season 5 of The Voice, as a team member of an American singer and lead vocalist of the pop-rock band Maroon 5, which released its fifth studio album (Roman numeral for five) with James Valentine as a guitarist.
|
{
"evidence": [
"Tessanne Chin",
"Adam Levine",
"V (Maroon 5 album)"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nTessanne Amanda Chin is a Canadian recording artist, best known for winning Season 5 of The Voice, as a team member of an American singer and lead vocalist of the pop-rock band Maroon 5, which released its fifth studio album (Roman numeral for five) with James Valentine as a guitarist.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Tessanne Chin",
"Adam Levine",
"V (Maroon 5 album)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3899,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6085
|
Three Flavours Cornetto trilogy: Shaun of the Dead is Simon John Pegg's first film.
|
{
"evidence": [
"Simon Pegg",
"Channel 4"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThree Flavours Cornetto trilogy: Shaun of the Dead is Simon John Pegg's first film.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Simon Pegg",
"Channel 4"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3900,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3184
|
Inside Man is Russell Gewirtz's most well-known film.
|
{
"evidence": [
"Inside Man",
"Denzel Washington"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nInside Man is Russell Gewirtz's most well-known film.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Inside Man",
"Denzel Washington"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3901,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_638
|
Saratoga is a 1937 American romantic comedy film starring an American actress known for her portrayal of "bad girl" characters.
|
{
"evidence": [
"Saratoga (film)",
"Jean Harlow"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nSaratoga is a 1937 American romantic comedy film starring an American actress known for her portrayal of \"bad girl\" characters.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Saratoga (film)",
"Jean Harlow"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3902,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2664
|
The Shape of Water was the first film written by Vanessa Taylor.
|
{
"evidence": [
"The Shape of Water (film)",
"Michael Shannon",
"Nocturnal Animals"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Shape of Water was the first film written by Vanessa Taylor.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Shape of Water (film)",
"Michael Shannon",
"Nocturnal Animals"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3903,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4744
|
Malaysia Airlines is the flag carrier airline of Malaysia and a member of an airline alliance that was founded on February 1st, 1999.
|
{
"evidence": [
"Malaysia Airlines",
"Oneworld"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nMalaysia Airlines is the flag carrier airline of Malaysia and a member of an airline alliance that was founded on February 1st, 1999.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Malaysia Airlines",
"Oneworld"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3904,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
43431
|
MG George E. Pickett, a career United States Army officer who was promoted as major general in the Confederate States Army during the American Civil War, commandeered five brigades in the Department of North Carolina and Southern Virginia.
|
{
"evidence": [
"George Pickett_sentence_2",
"George Pickett_sentence_135"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nMG George E. Pickett, a career United States Army officer who was promoted as major general in the Confederate States Army during the American Civil War, commandeered five brigades in the Department of North Carolina and Southern Virginia.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"George Pickett_sentence_2",
"George Pickett_sentence_135"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3905,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.394",
"value": "?search="
},
{
"operation": "search",
"time": "13.658",
"value": "George E. Pickett"
},
{
"operation": "Now on",
"time": "15.231",
"value": "George Pickett"
},
{
"operation": "Highlighting",
"time": "37.948",
"value": "George Pickett_sentence_2"
},
{
"operation": "hyperlink",
"time": "105.062",
"value": "3.4 North Carolina George Pickett_section_6"
},
{
"operation": "Highlighting",
"time": "112.253",
"value": "George Pickett_sentence_135"
},
{
"operation": "finish",
"time": "154.953",
"value": "finish"
}
]
}
|
exfever_528
|
Foo Fighters is an American band formed in 1994 whose instrumentation has centered on the electric guitar, usually as part of a group with the lowest-pitched member of the guitar family.
|
{
"evidence": [
"Foo Fighters",
"Rock music",
"Bass guitar"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nFoo Fighters is an American band formed in 1994 whose instrumentation has centered on the electric guitar, usually as part of a group with the lowest-pitched member of the guitar family.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Foo Fighters",
"Rock music",
"Bass guitar"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3906,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
81031
|
The 1936 International Society for Contemporary Music Festival showcased performers, including Leonor Meyer, Alice Frey, Mercedes Plantada, and Antonio Brosa from April 24-26
|
{
"evidence": [
"1936 International Society for Contemporary Music Festival_sentence_1"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe 1936 International Society for Contemporary Music Festival showcased performers, including Leonor Meyer, Alice Frey, Mercedes Plantada, and Antonio Brosa from April 24-26\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"1936 International Society for Contemporary Music Festival_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3907,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.79",
"value": "undefined"
},
{
"operation": "search",
"time": "11.828",
"value": "1936 International Society for Contemporary Music Festival"
},
{
"operation": "Now on",
"time": "12.934",
"value": "1936 International Society for Contemporary Music Festival"
},
{
"operation": "Highlighting",
"time": "33.15",
"value": "1936 International Society for Contemporary Music Festival_sentence_1"
},
{
"operation": "finish",
"time": "40.811",
"value": "finish"
}
]
}
|
exfever_8253
|
Boyd Holbrook is an American actor who starred as DEA agent Steve Murphy in the Netflix American crime drama television series created and produced by Chris Brancato.
|
{
"evidence": [
"Boyd Holbrook",
"Narcos"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBoyd Holbrook is an American actor who starred as DEA agent Steve Murphy in the Netflix American crime drama television series created and produced by Chris Brancato.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Boyd Holbrook",
"Narcos"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3908,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
1764
|
Lying on the border of the intermediate and dry zones, Kandalama is a village in Sri Lanka, in South Asia, nestled at the foot of Ereulagala and Dikkandahena in Sri Lanka and serves as a location of a 1700-year old irrigation tank.
|
{
"evidence": [
"Kandalama_sentence_6",
"Kandalama_sentence_0",
"Sri Lanka_sentence_3",
"Kandalama_sentence_5",
"Kandalama_sentence_4"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nLying on the border of the intermediate and dry zones, Kandalama is a village in Sri Lanka, in South Asia, nestled at the foot of Ereulagala and Dikkandahena in Sri Lanka and serves as a location of a 1700-year old irrigation tank.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Kandalama_sentence_6",
"Kandalama_sentence_0",
"Sri Lanka_sentence_3",
"Kandalama_sentence_5",
"Kandalama_sentence_4"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3909,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "search",
"time": "67.095",
"value": "kandalama"
},
{
"operation": "Page search",
"time": "160.76",
"value": "dry"
},
{
"operation": "Highlighting",
"time": "170.176",
"value": "Kandalama_sentence_6"
},
{
"operation": "Highlighting",
"time": "173.827",
"value": "Kandalama_sentence_0"
},
{
"operation": "hyperlink",
"time": "179.915",
"value": "Sri Lanka"
},
{
"operation": "Page search",
"time": "190.25",
"value": "south asia"
},
{
"operation": "Highlighting",
"time": "191.993",
"value": "Sri Lanka_sentence_3"
},
{
"operation": "back-button-clicked",
"time": "195.033",
"value": "back-button-clicked"
},
{
"operation": "Highlighting",
"time": "203.449",
"value": "Kandalama_sentence_5"
},
{
"operation": "Highlighting",
"time": "212.612",
"value": "Kandalama_sentence_4"
},
{
"operation": "Page search",
"time": "224.397",
"value": "south asia"
},
{
"operation": "finish",
"time": "230.479",
"value": "finish"
}
]
}
|
45628
|
Civil wrong or wrong is a cause of action under the law of the governing body, examples are Tort, breach of contract and breach of trust.
|
{
"evidence": [
"Civil wrong_sentence_1",
"Civil wrong_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nCivil wrong or wrong is a cause of action under the law of the governing body, examples are Tort, breach of contract and breach of trust.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Civil wrong_sentence_1",
"Civil wrong_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3910,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.884",
"value": "?search="
},
{
"operation": "search",
"time": "18.905",
"value": "Civil wrong or wrong"
},
{
"operation": "Now on",
"time": "20.381",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "33.756",
"value": "Civil wrong"
},
{
"operation": "Now on",
"time": "34.794",
"value": "Civil wrong"
},
{
"operation": "Highlighting",
"time": "48.08",
"value": "Civil wrong_sentence_1"
},
{
"operation": "Page search",
"time": "80.497",
"value": "tort"
},
{
"operation": "Highlighting",
"time": "83.531",
"value": "Civil wrong_sentence_2"
},
{
"operation": "finish",
"time": "112.056",
"value": "finish"
}
]
}
|
exfever_3440
|
Pulp Fiction is a 1994 American black comedy crime film starring an American singer best known for his supporting role as gang kingpin Marsellus Wallace in a 1994 American black comedy crime film.
|
{
"evidence": [
"Pulp Fiction",
"Ving Rhames",
"Pulp Fiction"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nPulp Fiction is a 1994 American black comedy crime film starring an American singer best known for his supporting role as gang kingpin Marsellus Wallace in a 1994 American black comedy crime film.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Pulp Fiction",
"Ving Rhames",
"Pulp Fiction"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3911,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7626
|
Pink is one of the major pop singers who won a Grammy.
|
{
"evidence": [
"Try (Pink song)",
"Greg Kurstin"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nPink is one of the major pop singers who won a Grammy.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Try (Pink song)",
"Greg Kurstin"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3912,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4735
|
The Hobbit is a children's fantasy novel by an American writer, poet, philologist, and academic, best known as the author of the high fantasy work The Lord of the Rings.
|
{
"evidence": [
"The Hobbit",
"J. R. R. Tolkien"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Hobbit is a children's fantasy novel by an American writer, poet, philologist, and academic, best known as the author of the high fantasy work The Lord of the Rings.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Hobbit",
"J. R. R. Tolkien"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3913,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
47334
|
James Joseph Patterson was the son of Joseph Medill Patterson, the founder of the New York Daily News which was the first U.S. daily newspaper to be sold by newsies, and was the eleventh-highest circulated newspaper in the United States as of 2019
|
{
"evidence": [
"James Joseph Patterson_sentence_3",
"New York Daily News_sentence_4"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJames Joseph Patterson was the son of Joseph Medill Patterson, the founder of the New York Daily News which was the first U.S. daily newspaper to be sold by newsies, and was the eleventh-highest circulated newspaper in the United States as of 2019\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"James Joseph Patterson_sentence_3",
"New York Daily News_sentence_4"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3914,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.689",
"value": "?search="
},
{
"operation": "search",
"time": "7.043",
"value": "James Joseph Patterson"
},
{
"operation": "Now on",
"time": "8.225",
"value": "James Joseph Patterson"
},
{
"operation": "Highlighting",
"time": "16.035",
"value": "James Joseph Patterson_sentence_3"
},
{
"operation": "hyperlink",
"time": "19.466",
"value": "New York Daily News"
},
{
"operation": "Now on",
"time": "20.649",
"value": "New York Daily News"
},
{
"operation": "search",
"time": "63.046",
"value": "newsies"
},
{
"operation": "Now on",
"time": "64.023",
"value": "Newsies"
},
{
"operation": "back-button-clicked",
"time": "79.386",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "79.54",
"value": "New York Daily News"
},
{
"operation": "Highlighting",
"time": "132.89",
"value": "New York Daily News_sentence_4"
},
{
"operation": "hyperlink",
"time": "153.421",
"value": "Media in New York City"
},
{
"operation": "Now on",
"time": "155.242",
"value": "Media in New York City"
},
{
"operation": "back-button-clicked",
"time": "307.031",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "307.226",
"value": "New York Daily News"
},
{
"operation": "search",
"time": "339.247",
"value": "distribution of New York Daily News"
},
{
"operation": "Now on",
"time": "342.41",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "354.831",
"value": "List of tabloids"
},
{
"operation": "Now on",
"time": "355.719",
"value": "Tabloid (newspaper format)"
},
{
"operation": "hyperlink",
"time": "368.174",
"value": "New York Daily News"
},
{
"operation": "Now on",
"time": "368.543",
"value": "New York Daily News"
},
{
"operation": "hyperlink",
"time": "387.805",
"value": "tabloid format"
},
{
"operation": "Now on",
"time": "388.595",
"value": "Tabloid (newspaper format)"
},
{
"operation": "back-button-clicked",
"time": "400.857",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "400.925",
"value": "New York Daily News"
},
{
"operation": "hyperlink",
"time": "405.378",
"value": "Daily newspaper"
},
{
"operation": "Now on",
"time": "407.14",
"value": "Newspaper"
},
{
"operation": "back-button-clicked",
"time": "448.636",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "448.707",
"value": "New York Daily News"
},
{
"operation": "finish",
"time": "479.021",
"value": "finish"
}
]
}
|
47761
|
Savage, an English nonconformist minister and dissenting tutor, first thought of medicine, and spent a year or two with his Uncle Toulmin, an apothecary, in Old Gravel Lane, Wapping; then through the influence of Isaac Watts, he entered the Fund Academy.
|
{
"evidence": [
"Samuel Morton Savage_sentence_0",
"Samuel Morton Savage_sentence_4",
"Samuel Morton Savage_sentence_5"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nSavage, an English nonconformist minister and dissenting tutor, first thought of medicine, and spent a year or two with his Uncle Toulmin, an apothecary, in Old Gravel Lane, Wapping; then through the influence of Isaac Watts, he entered the Fund Academy.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Samuel Morton Savage_sentence_0",
"Samuel Morton Savage_sentence_4",
"Samuel Morton Savage_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3915,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.001",
"value": "undefined"
},
{
"operation": "search",
"time": "0.012",
"value": "Savage"
},
{
"operation": "Now on",
"time": "0.013",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "0.016",
"value": "Savage"
},
{
"operation": "Now on",
"time": "0.018",
"value": "Savage"
},
{
"operation": "search",
"time": "0.097",
"value": "English nonconformist minister and dissenting tutor"
},
{
"operation": "Now on",
"time": "0.099",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "0.112",
"value": "Samuel Morton Savage"
},
{
"operation": "Now on",
"time": "0.113",
"value": "Samuel Morton Savage"
},
{
"operation": "Highlighting",
"time": "0.115",
"value": "Samuel Morton Savage_sentence_0"
},
{
"operation": "Page search",
"time": "0.139",
"value": "medicine"
},
{
"operation": "Page search",
"time": "0.139",
"value": "medicine"
},
{
"operation": "Page search",
"time": "0.139",
"value": "medicine"
},
{
"operation": "Highlighting",
"time": "0.143",
"value": "Samuel Morton Savage_sentence_4"
},
{
"operation": "Highlighting",
"time": "0.158",
"value": "Samuel Morton Savage_sentence_5"
},
{
"operation": "finish",
"time": "0.177",
"value": "finish"
},
{
"operation": "Now on",
"time": "1620936409.668",
"value": "Samuel Morton Savage"
},
{
"operation": "finish",
"time": "1620936411.552",
"value": "finish"
}
]
}
|
24216
|
English author, Henry Arthur Bright, was related to Richard Monckton Milnes, 1st Baron Houghton, who was also an English poet, patron of literature.
|
{
"evidence": [
"Henry Arthur Bright_sentence_3",
"Richard Monckton Milnes, 1st Baron Houghton_sentence_0"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nEnglish author, Henry Arthur Bright, was related to Richard Monckton Milnes, 1st Baron Houghton, who was also an English poet, patron of literature.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Henry Arthur Bright_sentence_3",
"Richard Monckton Milnes, 1st Baron Houghton_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3916,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.485",
"value": "?search="
},
{
"operation": "search",
"time": "41.493",
"value": "Henry Arthur Bright"
},
{
"operation": "Now on",
"time": "42.034",
"value": "Henry Arthur Bright"
},
{
"operation": "Highlighting",
"time": "52.452",
"value": "Henry Arthur Bright_sentence_3"
},
{
"operation": "hyperlink",
"time": "55.948",
"value": "Richard Monckton Milnes, 1st Baron Houghton"
},
{
"operation": "Now on",
"time": "56.354",
"value": "Richard Monckton Milnes, 1st Baron Houghton"
},
{
"operation": "Highlighting",
"time": "59.581",
"value": "Richard Monckton Milnes, 1st Baron Houghton_sentence_0"
},
{
"operation": "finish",
"time": "76.388",
"value": "finish"
}
]
}
|
63275
|
Francisco Trincão only played baseball in his entire sports career life.
|
{
"evidence": [
"Francisco Trincão_sentence_1"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nFrancisco Trincão only played baseball in his entire sports career life.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Francisco Trincão_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3917,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.114",
"value": "?search="
},
{
"operation": "search",
"time": "9.748",
"value": "Francisco Trincão"
},
{
"operation": "Now on",
"time": "11.198",
"value": "Francisco Trincão"
},
{
"operation": "Highlighting",
"time": "15.668",
"value": "Francisco Trincão_sentence_1"
},
{
"operation": "finish",
"time": "41.089",
"value": "finish"
}
]
}
|
82206
|
Richard Corish ran as a member of the Labour Party at the 1922 general election—an Irish general election that served as the election to the Third Dáil—and later on served in the Dáil.
|
{
"evidence": [
"Richard Corish_sentence_5",
"1922 Irish general election_sentence_1",
"Richard Corish_sentence_6"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nRichard Corish ran as a member of the Labour Party at the 1922 general election—an Irish general election that served as the election to the Third Dáil—and later on served in the Dáil.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Richard Corish_sentence_5",
"1922 Irish general election_sentence_1",
"Richard Corish_sentence_6"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3918,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.21",
"value": "?search="
},
{
"operation": "search",
"time": "48.332",
"value": "Richard Corish"
},
{
"operation": "Now on",
"time": "49.234",
"value": "Richard Corish"
},
{
"operation": "Highlighting",
"time": "99.458",
"value": "Richard Corish_sentence_5"
},
{
"operation": "hyperlink",
"time": "111.8",
"value": "1922 general election"
},
{
"operation": "Now on",
"time": "112.328",
"value": "1922 Irish general election"
},
{
"operation": "Highlighting",
"time": "193.033",
"value": "1922 Irish general election_sentence_0"
},
{
"operation": "Highlighting",
"time": "194.104",
"value": "1922 Irish general election_sentence_1"
},
{
"operation": "back-button-clicked",
"time": "201.553",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "201.759",
"value": "Richard Corish"
},
{
"operation": "Highlighting",
"time": "260.447",
"value": "Richard Corish_sentence_6"
},
{
"operation": "hyperlink",
"time": "331.496",
"value": "1922 general election"
},
{
"operation": "Now on",
"time": "332.251",
"value": "1922 Irish general election"
},
{
"operation": "Highlighting",
"time": "334.256",
"value": "1922 Irish general election_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "334.847",
"value": "1922 Irish general election_sentence_0"
},
{
"operation": "back-button-clicked",
"time": "345.809",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "345.992",
"value": "Richard Corish"
},
{
"operation": "Highlighting deleted",
"time": "349.231",
"value": "Richard Corish_sentence_6"
},
{
"operation": "Highlighting",
"time": "370.239",
"value": "Richard Corish_sentence_6"
},
{
"operation": "finish",
"time": "376.103",
"value": "finish"
}
]
}
|
exfever_4411
|
The Enough Project is a non-profit organization that conducts research in several conflict areas in Africa that were controlled by a Christian extremist and a terrorist organization founded in 1987.
|
{
"evidence": [
"Enough Project",
"Lord's Resistance Army"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Enough Project is a non-profit organization that conducts research in several conflict areas in Africa that were controlled by a Christian extremist and a terrorist organization founded in 1987.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Enough Project",
"Lord's Resistance Army"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3919,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3659
|
Walter Matthau was an American actor known for his film roles in an American sports comedy film starring an American actress and author.
|
{
"evidence": [
"Walter Matthau",
"The Bad News Bears",
"Tatum O'Neal"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nWalter Matthau was an American actor known for his film roles in an American sports comedy film starring an American actress and author.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Walter Matthau",
"The Bad News Bears",
"Tatum O'Neal"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3920,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4758
|
Hungary is a landlocked country on a continent that is bordered by the smallest and shallowest of the world's five major oceans.
|
{
"evidence": [
"Hungary",
"Europe",
"Arctic Ocean"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nHungary is a landlocked country on a continent that is bordered by the smallest and shallowest of the world's five major oceans.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Hungary",
"Europe",
"Arctic Ocean"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3921,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_715
|
Houston is located in Southeast Texas near Galveston Bay and the Gulf of Mexico, which is the seat and largest city of the County in the U.S. state of Alaska.
|
{
"evidence": [
"Houston",
"Harris County, Texas"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nHouston is located in Southeast Texas near Galveston Bay and the Gulf of Mexico, which is the seat and largest city of the County in the U.S. state of Alaska.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Houston",
"Harris County, Texas"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3922,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
81836
|
Nmperign is an experimental jazz group formed in Bosten In 1998 and only releases singles.
|
{
"evidence": [
"Nmperign_sentence_5"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nNmperign is an experimental jazz group formed in Bosten In 1998 and only releases singles.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Nmperign_sentence_5"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3923,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.83",
"value": "?search="
},
{
"operation": "search",
"time": "72.526",
"value": "Nmperign"
},
{
"operation": "Now on",
"time": "73.367",
"value": "Nmperign"
},
{
"operation": "Highlighting",
"time": "79.742",
"value": "Nmperign_sentence_0"
},
{
"operation": "Highlighting",
"time": "91.815",
"value": "Nmperign_sentence_5"
},
{
"operation": "Highlighting deleted",
"time": "94.349",
"value": "Nmperign_sentence_0"
},
{
"operation": "finish",
"time": "102.887",
"value": "finish"
}
]
}
|
38473
|
Jane Weinzapfel, an Polish engineer and architect, was one of the creators of the architecture firm Leers Weinzapfel Associates.
|
{
"evidence": [
"Jane Weinzapfel_sentence_0"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJane Weinzapfel, an Polish engineer and architect, was one of the creators of the architecture firm Leers Weinzapfel Associates.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Jane Weinzapfel_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3924,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.246",
"value": "?search="
},
{
"operation": "search",
"time": "13.489",
"value": "Leers Weinzapfel Associates"
},
{
"operation": "Now on",
"time": "15.012",
"value": "Leers Weinzapfel Associates"
},
{
"operation": "Highlighting",
"time": "24.397",
"value": "Leers Weinzapfel Associates_sentence_1"
},
{
"operation": "Highlighting",
"time": "24.399",
"value": "Leers Weinzapfel Associates_title"
},
{
"operation": "hyperlink",
"time": "26.419",
"value": "Jane Weinzapfel"
},
{
"operation": "Now on",
"time": "27.12",
"value": "Jane Weinzapfel"
},
{
"operation": "Highlighting",
"time": "36.491",
"value": "Jane Weinzapfel_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "42.307",
"value": "Jane Weinzapfel_sentence_1"
},
{
"operation": "Highlighting",
"time": "43.075",
"value": "Jane Weinzapfel_sentence_0"
},
{
"operation": "finish",
"time": "55.284",
"value": "finish"
}
]
}
|
45094
|
Arthur Terry came to Barcelona for the first time in 1949, he taught Hispanic languages and was appointed literature professor at the University of Essex.
|
{
"evidence": [
"Arthur Terry_sentence_4",
"Arthur Terry_sentence_5",
"Arthur Terry_sentence_6"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nArthur Terry came to Barcelona for the first time in 1949, he taught Hispanic languages and was appointed literature professor at the University of Essex.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Arthur Terry_sentence_4",
"Arthur Terry_sentence_5",
"Arthur Terry_sentence_6"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3925,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.847",
"value": "?search="
},
{
"operation": "search",
"time": "13.643",
"value": "Arthur Terry"
},
{
"operation": "Now on",
"time": "14.845",
"value": "Arthur Terry"
},
{
"operation": "Highlighting",
"time": "24.601",
"value": "Arthur Terry_sentence_4"
},
{
"operation": "Highlighting",
"time": "31.235",
"value": "Arthur Terry_sentence_5"
},
{
"operation": "Highlighting",
"time": "35.613",
"value": "Arthur Terry_sentence_6"
},
{
"operation": "finish",
"time": "39.735",
"value": "finish"
}
]
}
|
exfever_4358
|
The human condition is all of the characteristics and key events that compose the essentials of animal existence, which has been pondered and analyzed from many perspectives, including the study of general and fundamental questions.
|
{
"evidence": [
"Human condition",
"Philosophy"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe human condition is all of the characteristics and key events that compose the essentials of animal existence, which has been pondered and analyzed from many perspectives, including the study of general and fundamental questions.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Human condition",
"Philosophy"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3926,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
57735
|
Katharine Banham was appointed as a Senior Psychologist to the North Carolina Board of Public Welfare four years after becoming an associate professor at Duke University.
|
{
"evidence": [
"Katharine Banham_sentence_7",
"Katharine Banham_sentence_14"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nKatharine Banham was appointed as a Senior Psychologist to the North Carolina Board of Public Welfare four years after becoming an associate professor at Duke University.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Katharine Banham_sentence_7",
"Katharine Banham_sentence_14"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3927,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.274",
"value": "?search="
},
{
"operation": "search",
"time": "13.712",
"value": "Katharine Banham"
},
{
"operation": "Now on",
"time": "14.638",
"value": "Katharine Banham"
},
{
"operation": "hyperlink",
"time": "79.182",
"value": "Cerebral Palsy"
},
{
"operation": "Now on",
"time": "81.403",
"value": "Katharine Banham"
},
{
"operation": "Highlighting",
"time": "82.538",
"value": "Katharine Banham_sentence_7"
},
{
"operation": "Highlighting",
"time": "115.138",
"value": "Katharine Banham_sentence_14"
},
{
"operation": "finish",
"time": "138.983",
"value": "finish"
}
]
}
|
36803
|
A constitutional crisis occurs when political parties willfully violate a constitution's law, as was demonstrated during the XYZ Affair.
|
{
"evidence": [
"Constitutional crisis_sentence_0"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nA constitutional crisis occurs when political parties willfully violate a constitution's law, as was demonstrated during the XYZ Affair.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Constitutional crisis_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3928,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.153",
"value": "?search="
},
{
"operation": "search",
"time": "15.459",
"value": "Constitutional crisis"
},
{
"operation": "Now on",
"time": "16.507",
"value": "Constitutional crisis"
},
{
"operation": "search",
"time": "69.745",
"value": "XYZ Affair"
},
{
"operation": "Now on",
"time": "70.716",
"value": "XYZ Affair"
},
{
"operation": "Page search",
"time": "99.576",
"value": "constitutional"
},
{
"operation": "back-button-clicked",
"time": "141.203",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "141.473",
"value": "Constitutional crisis"
},
{
"operation": "Highlighting",
"time": "156.095",
"value": "Constitutional crisis_sentence_0"
},
{
"operation": "finish",
"time": "177.735",
"value": "finish"
}
]
}
|
14027
|
Winter Saloon also known as Harm's Bar is a historic structure that was placed on the National Register of Historic Places on January 4, 1980.
|
{
"evidence": [
"Winter Saloon_sentence_0",
"Winter Saloon_sentence_1"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nWinter Saloon also known as Harm's Bar is a historic structure that was placed on the National Register of Historic Places on January 4, 1980.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Winter Saloon_sentence_0",
"Winter Saloon_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3929,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.444",
"value": "?search="
},
{
"operation": "search",
"time": "14.992",
"value": "Winter Saloon"
},
{
"operation": "Now on",
"time": "15.764",
"value": "Winter Saloon"
},
{
"operation": "Highlighting",
"time": "21.368",
"value": "Winter Saloon_sentence_0"
},
{
"operation": "Highlighting",
"time": "25.895",
"value": "Winter Saloon_header_cell_0_7_0"
},
{
"operation": "Highlighting",
"time": "26.959",
"value": "Winter Saloon_cell_0_7_1"
},
{
"operation": "Highlighting",
"time": "26.963",
"value": "Winter Saloon_header_cell_0_7_0"
},
{
"operation": "Highlighting deleted",
"time": "32.327",
"value": "Winter Saloon_cell_0_7_1"
},
{
"operation": "Highlighting deleted",
"time": "32.33",
"value": "Winter Saloon_header_cell_0_7_0"
},
{
"operation": "Highlighting",
"time": "33.807",
"value": "Winter Saloon_sentence_1"
},
{
"operation": "finish",
"time": "42.727",
"value": "finish"
}
]
}
|
42684
|
Emmy award-winning producer Steve Benen served as lead blogger of Political Animal blog and published The Carpetbagger Report.
|
{
"evidence": [
"Steve Benen_sentence_0",
"Steve Benen_sentence_6",
"Steve Benen_sentence_7"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nEmmy award-winning producer Steve Benen served as lead blogger of Political Animal blog and published The Carpetbagger Report.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Steve Benen_sentence_0",
"Steve Benen_sentence_6",
"Steve Benen_sentence_7"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3930,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.296",
"value": "?search="
},
{
"operation": "search",
"time": "12.615",
"value": "Steve Benen"
},
{
"operation": "Now on",
"time": "13.837",
"value": "Steve Benen"
},
{
"operation": "Highlighting",
"time": "17.772",
"value": "Steve Benen_sentence_0"
},
{
"operation": "Highlighting",
"time": "32.108",
"value": "Steve Benen_sentence_6"
},
{
"operation": "Highlighting",
"time": "41.396",
"value": "Steve Benen_sentence_7"
},
{
"operation": "finish",
"time": "47.532",
"value": "finish"
}
]
}
|
exfever_2995
|
Adventure Time is an American fantasy television series starring an actor known for his role as Dr. Drakken in Kim Possible.
|
{
"evidence": [
"Adventure Time",
"John DiMaggio"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nAdventure Time is an American fantasy television series starring an actor known for his role as Dr. Drakken in Kim Possible.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Adventure Time",
"John DiMaggio"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3931,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5335
|
Lewis Friedman is an American screenwriter whose credits include the 1998 drama film starring an American actor, animator, writer, producer, and composer.
|
{
"evidence": [
"Lewis Friedman",
"BASEketball",
"Matt Stone"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nLewis Friedman is an American screenwriter whose credits include the 1998 drama film starring an American actor, animator, writer, producer, and composer.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Lewis Friedman",
"BASEketball",
"Matt Stone"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3932,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
75575
|
Seven people ran in the primary for the 1922 US Senate election in Texas with Earle B. Mayfield, a lawyer and Democratic Party politician from Overton, Texas, receiving 26.78% of the votes.
|
{
"evidence": [
"Earle Bradford Mayfield_sentence_9"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nSeven people ran in the primary for the 1922 US Senate election in Texas with Earle B. Mayfield, a lawyer and Democratic Party politician from Overton, Texas, receiving 26.78% of the votes.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Earle Bradford Mayfield_sentence_9"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3933,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.254",
"value": "undefined"
},
{
"operation": "search",
"time": "43.997",
"value": "Primary for the 1922 US Senate election in Texas"
},
{
"operation": "Now on",
"time": "46.058",
"value": "undefined"
},
{
"operation": "search",
"time": "70.881",
"value": "1922 elections in Texas"
},
{
"operation": "Now on",
"time": "71.761",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "87.081",
"value": "1922 United States Senate elections"
},
{
"operation": "Now on",
"time": "87.846",
"value": "1922 United States Senate elections"
},
{
"operation": "hyperlink",
"time": "105.352",
"value": "30 Texas 1922 United States Senate elections_section_38"
},
{
"operation": "Highlighting",
"time": "107.489",
"value": "1922 United States Senate elections_cell_33_1_1"
},
{
"operation": "Highlighting",
"time": "107.49",
"value": "1922 United States Senate elections_title"
},
{
"operation": "Highlighting",
"time": "107.49",
"value": "1922 United States Senate elections_section_38"
},
{
"operation": "Highlighting",
"time": "107.491",
"value": "1922 United States Senate elections_header_cell_33_0_0"
},
{
"operation": "Highlighting",
"time": "109.104",
"value": "1922 United States Senate elections_cell_33_1_2"
},
{
"operation": "Highlighting",
"time": "109.106",
"value": "1922 United States Senate elections_header_cell_33_0_2"
},
{
"operation": "hyperlink",
"time": "115.424",
"value": "Earle Bradford Mayfield"
},
{
"operation": "Now on",
"time": "115.729",
"value": "Earle Bradford Mayfield"
},
{
"operation": "Highlighting",
"time": "119.336",
"value": "Earle Bradford Mayfield_sentence_2"
},
{
"operation": "Highlighting",
"time": "152.415",
"value": "Earle Bradford Mayfield_sentence_9"
},
{
"operation": "hyperlink",
"time": "168.686",
"value": "Democratic Party"
},
{
"operation": "Now on",
"time": "170.15",
"value": "Earle Bradford Mayfield"
},
{
"operation": "Highlighting deleted",
"time": "171.294",
"value": "Earle Bradford Mayfield_sentence_2"
},
{
"operation": "finish",
"time": "191.502",
"value": "finish"
}
]
}
|
exfever_2734
|
Salman Khan is an Indian actor who has received awards that honor the film industry of India by an Indian English-language fortnightly magazine published by Worldwide Media.
|
{
"evidence": [
"Salman Khan",
"Filmfare Awards",
"Filmfare"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nSalman Khan is an Indian actor who has received awards that honor the film industry of India by an Indian English-language fortnightly magazine published by Worldwide Media.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Salman Khan",
"Filmfare Awards",
"Filmfare"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3934,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5390
|
Jacqueline Kennedy Onassis was an American socialite who was the wife of an American politician born into a prominent American political family.
|
{
"evidence": [
"Jacqueline Kennedy Onassis",
"John F. Kennedy",
"Kennedy family"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJacqueline Kennedy Onassis was an American socialite who was the wife of an American politician born into a prominent American political family.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Jacqueline Kennedy Onassis",
"John F. Kennedy",
"Kennedy family"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3935,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5657
|
San Junipero is revealed to be a simulated reality where the deceased can live and the elderly can visit.
|
{
"evidence": [
"San Junipero",
"Black Mirror"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nSan Junipero is revealed to be a simulated reality where the deceased can live and the elderly can visit.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"San Junipero",
"Black Mirror"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3936,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
52655
|
Grant Lightbown retired in 1991.
|
{
"evidence": [
"Grant Lightbown_sentence_1"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nGrant Lightbown retired in 1991.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Grant Lightbown_sentence_1"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3937,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.307",
"value": "?search="
},
{
"operation": "search",
"time": "24.631",
"value": "Grant Lightbown"
},
{
"operation": "Now on",
"time": "25.624",
"value": "Grant Lightbown"
},
{
"operation": "Highlighting",
"time": "287.9",
"value": "Grant Lightbown_sentence_1"
},
{
"operation": "finish",
"time": "355.154",
"value": "finish"
}
]
}
|
exfever_3906
|
Gaius Julius Caesar was a general and statesman who became an extraordinary magistrate in the Roman Republic, endowed with full authority to resolve some specific problem to which he had been assigned.
|
{
"evidence": [
"Julius Caesar",
"Roman dictator"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nGaius Julius Caesar was a general and statesman who became an extraordinary magistrate in the Roman Republic, endowed with full authority to resolve some specific problem to which he had been assigned.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Julius Caesar",
"Roman dictator"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3938,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
8605
|
A notable exception of the Modern Greek theatre is the comedy Babylonia by Galileo Galilei in 1836 and the Royal Theatre was founded in 1880 in Paris and re-founded in 1932 as Arena Theatre.
|
{
"evidence": [
"Modern Greek theatre_sentence_15"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nA notable exception of the Modern Greek theatre is the comedy Babylonia by Galileo Galilei in 1836 and the Royal Theatre was founded in 1880 in Paris and re-founded in 1932 as Arena Theatre.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Modern Greek theatre_sentence_15"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3939,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.283",
"value": "?search="
},
{
"operation": "search",
"time": "44.714",
"value": "Modern Greek theatre"
},
{
"operation": "Now on",
"time": "46.086",
"value": "Modern Greek theatre"
},
{
"operation": "Page search",
"time": "50.549",
"value": "babylonia"
},
{
"operation": "Highlighting",
"time": "55.421",
"value": "Modern Greek theatre_sentence_15"
},
{
"operation": "search",
"time": "72.993",
"value": "Royal Theatre"
},
{
"operation": "Now on",
"time": "74.078",
"value": "Royal Theatre"
},
{
"operation": "search",
"time": "97.973",
"value": "Contains...arena theatre"
},
{
"operation": "Now on",
"time": "99.081",
"value": "undefined"
},
{
"operation": "search",
"time": "123.751",
"value": "Royal Theatre"
},
{
"operation": "Now on",
"time": "124.431",
"value": "Royal Theatre"
},
{
"operation": "hyperlink",
"time": "128.572",
"value": "1 Venues Royal Theatre_section_0"
},
{
"operation": "Highlighting",
"time": "138.454",
"value": "Royal Theatre_item_9_14"
},
{
"operation": "Highlighting",
"time": "138.455",
"value": "Royal Theatre_title"
},
{
"operation": "Highlighting",
"time": "138.456",
"value": "Royal Theatre_section_0"
},
{
"operation": "Highlighting",
"time": "138.456",
"value": "Royal Theatre_section_10"
},
{
"operation": "hyperlink",
"time": "139.896",
"value": "Royal National Theatre"
},
{
"operation": "Now on",
"time": "141.639",
"value": "Royal National Theatre"
},
{
"operation": "back-button-clicked",
"time": "170.102",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "170.25",
"value": "Royal Theatre#Venues Royal Theatre section 0"
},
{
"operation": "Highlighting",
"time": "173.599",
"value": "Royal Theatre_sentence_0"
},
{
"operation": "Highlighting",
"time": "173.601",
"value": "Royal Theatre_title"
},
{
"operation": "hyperlink",
"time": "183.316",
"value": "2 Other uses Royal Theatre_section_12"
},
{
"operation": "Highlighting deleted",
"time": "222.654",
"value": "Royal Theatre_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "222.656",
"value": "Royal Theatre_title"
},
{
"operation": "finish",
"time": "233.634",
"value": "finish"
}
]
}
|
exfever_1499
|
Claire Danes got married to Hugh Dancy.
|
{
"evidence": [
"Claire Danes",
"Screen Actors Guild Award"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nClaire Danes got married to Hugh Dancy.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Claire Danes",
"Screen Actors Guild Award"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3940,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
51547
|
David Honey was elected to the Western Australian Legislative Assembly after the resignation of Colin Barnett, the 29th Premier of Western Australia.
|
{
"evidence": [
"David Honey_sentence_2",
"Colin Barnett_sentence_0"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nDavid Honey was elected to the Western Australian Legislative Assembly after the resignation of Colin Barnett, the 29th Premier of Western Australia.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"David Honey_sentence_2",
"Colin Barnett_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3941,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.8",
"value": "?search="
},
{
"operation": "search",
"time": "9.981",
"value": "David Honey"
},
{
"operation": "Now on",
"time": "10.814",
"value": "David Honey"
},
{
"operation": "Highlighting",
"time": "23.493",
"value": "David Honey_sentence_2"
},
{
"operation": "hyperlink",
"time": "48.785",
"value": "Colin Barnett"
},
{
"operation": "Now on",
"time": "50.233",
"value": "Colin Barnett"
},
{
"operation": "Highlighting",
"time": "51.711",
"value": "Colin Barnett_sentence_0"
},
{
"operation": "finish",
"time": "76.091",
"value": "finish"
}
]
}
|
exfever_3260
|
U2 are an Irish rock band whose members include an Irish musician who resides in County Dublin, in a republic country in north-western Africa.
|
{
"evidence": [
"U2",
"Adam Clayton",
"Republic of Ireland"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nU2 are an Irish rock band whose members include an Irish musician who resides in County Dublin, in a republic country in north-western Africa.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"U2",
"Adam Clayton",
"Republic of Ireland"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3942,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7989
|
Mexico is a country in the southern portion of North America which is the 9th-most-populous country and has the most speakers of a Romance language of the Indo-European language family which is the world's fourth-most spoken language overall after a West Germanic language of the Indo-European language family, originally spoken by the inhabitants of early medieval England.
|
{
"evidence": [
"Mexico",
"Spanish language",
"English language"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nMexico is a country in the southern portion of North America which is the 9th-most-populous country and has the most speakers of a Romance language of the Indo-European language family which is the world's fourth-most spoken language overall after a West Germanic language of the Indo-European language family, originally spoken by the inhabitants of early medieval England.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Mexico",
"Spanish language",
"English language"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3943,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
34386
|
Hector Aldana, known for his cardio and grinding fighting style, has historically gone to the scorecards for a decision throughout his career.
|
{
"evidence": [
"Hector Aldana_sentence_0"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nHector Aldana, known for his cardio and grinding fighting style, has historically gone to the scorecards for a decision throughout his career.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Hector Aldana_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3944,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.269",
"value": "?search="
},
{
"operation": "search",
"time": "7.179",
"value": "Hector Aldana"
},
{
"operation": "Now on",
"time": "8.985",
"value": "Hector Aldana"
},
{
"operation": "Page search",
"time": "30.829",
"value": "cardio"
},
{
"operation": "hyperlink",
"time": "42.94",
"value": "mixed martial artist"
},
{
"operation": "Now on",
"time": "44.608",
"value": "Mixed martial arts"
},
{
"operation": "back-button-clicked",
"time": "56.923",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "57.057",
"value": "Hector Aldana"
},
{
"operation": "Highlighting",
"time": "92.947",
"value": "Hector Aldana_sentence_0"
},
{
"operation": "finish",
"time": "100.408",
"value": "finish"
}
]
}
|
exfever_4070
|
Jacob Benjamin Gyllenhaal began acting as a child with roles in his father's film, where its feature was co-produced by Gramercy Pictures and an American film production company founded by director and producer Steven Spielberg.
|
{
"evidence": [
"Jake Gyllenhaal",
"A Dangerous Woman (1993 film)",
"Amblin Entertainment"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJacob Benjamin Gyllenhaal began acting as a child with roles in his father's film, where its feature was co-produced by Gramercy Pictures and an American film production company founded by director and producer Steven Spielberg.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Jake Gyllenhaal",
"A Dangerous Woman (1993 film)",
"Amblin Entertainment"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3945,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4388
|
Saratoga is a 1937 American romantic drama film starring an American actress known for her portrayal of "bad girl" characters.
|
{
"evidence": [
"Saratoga (film)",
"Jean Harlow"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nSaratoga is a 1937 American romantic drama film starring an American actress known for her portrayal of \"bad girl\" characters.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Saratoga (film)",
"Jean Harlow"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3946,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
88568
|
DJ Mamy Rock claimed to be engaged to Polly Brown, the songwriting/production team of Gerry Shury and Ron Roker admired her and her album reached Top Ten.
|
{
"evidence": [
"Polly Brown_sentence_4"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nDJ Mamy Rock claimed to be engaged to Polly Brown, the songwriting/production team of Gerry Shury and Ron Roker admired her and her album reached Top Ten.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Polly Brown_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3947,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.492",
"value": "?search="
},
{
"operation": "search",
"time": "9.97",
"value": "Contains...DJ Mamy Rock"
},
{
"operation": "Now on",
"time": "10.769",
"value": "undefined"
},
{
"operation": "search",
"time": "29.648",
"value": "Mamy Rock"
},
{
"operation": "Now on",
"time": "30.438",
"value": "undefined"
},
{
"operation": "search",
"time": "53.107",
"value": "Polly Brown"
},
{
"operation": "Now on",
"time": "54.051",
"value": "Polly Brown"
},
{
"operation": "Page search",
"time": "61.723",
"value": "rock"
},
{
"operation": "Page search",
"time": "76.381",
"value": "engage"
},
{
"operation": "Highlighting",
"time": "79.153",
"value": "Polly Brown_sentence_4"
},
{
"operation": "hyperlink",
"time": "80.497",
"value": "Jimmy Savile"
},
{
"operation": "Now on",
"time": "81.655",
"value": "Jimmy Savile"
},
{
"operation": "Page search",
"time": "105.543",
"value": "rock"
},
{
"operation": "back-button-clicked",
"time": "114",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "114.109",
"value": "Polly Brown"
},
{
"operation": "Highlighting",
"time": "126.576",
"value": "Polly Brown_sentence_5"
},
{
"operation": "Page search",
"time": "161.711",
"value": "top ten"
},
{
"operation": "Highlighting",
"time": "174.336",
"value": "Polly Brown_sentence_7"
},
{
"operation": "Highlighting deleted",
"time": "247.729",
"value": "Polly Brown_sentence_7"
},
{
"operation": "Highlighting deleted",
"time": "254.289",
"value": "Polly Brown_sentence_5"
},
{
"operation": "finish",
"time": "265.695",
"value": "finish"
}
]
}
|
77400
|
Conall Crandomna, king of Dál Riata from about 650 until 660, died in 659 or 660 and was subsequently succeeded by his nephew Domangart mac Domnaill.
|
{
"evidence": [
"Conall Crandomna_sentence_0",
"Conall Crandomna_sentence_5",
"Conall Crandomna_sentence_6"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nConall Crandomna, king of Dál Riata from about 650 until 660, died in 659 or 660 and was subsequently succeeded by his nephew Domangart mac Domnaill.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Conall Crandomna_sentence_0",
"Conall Crandomna_sentence_5",
"Conall Crandomna_sentence_6"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3948,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.116",
"value": "?search="
},
{
"operation": "search",
"time": "11.887",
"value": "Conall Crandomna"
},
{
"operation": "Now on",
"time": "12.72",
"value": "Conall Crandomna"
},
{
"operation": "Highlighting",
"time": "17.134",
"value": "Conall Crandomna_sentence_0"
},
{
"operation": "Highlighting",
"time": "21.894",
"value": "Conall Crandomna_sentence_5"
},
{
"operation": "Highlighting",
"time": "26.886",
"value": "Conall Crandomna_sentence_6"
},
{
"operation": "finish",
"time": "37.318",
"value": "finish"
}
]
}
|
exfever_1185
|
Broadchurch is a British crime drama television series that is set in Broadchurch, a fictional English town on the coast of a county in the South West of England on the English Channel coast.
|
{
"evidence": [
"Broadchurch",
"Dorset"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBroadchurch is a British crime drama television series that is set in Broadchurch, a fictional English town on the coast of a county in the South West of England on the English Channel coast.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Broadchurch",
"Dorset"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3949,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1477
|
"Just the Way You Are" is the debut solo single by Bruno Mars and written by American songwriting and record production team established in a state in the Western United States with over 39.5 billion residents.
|
{
"evidence": [
"Just the Way You Are (Bruno Mars song)",
"The Smeezingtons",
"California"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\n\"Just the Way You Are\" is the debut solo single by Bruno Mars and written by American songwriting and record production team established in a state in the Western United States with over 39.5 billion residents.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Just the Way You Are (Bruno Mars song)",
"The Smeezingtons",
"California"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3950,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1404
|
George W. Bush is a Canadian politician and member of an American dynastic family prominent in Russian politics, news, sports, entertainment, and business.
|
{
"evidence": [
"George W. Bush",
"Bush family"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nGeorge W. Bush is a Canadian politician and member of an American dynastic family prominent in Russian politics, news, sports, entertainment, and business.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"George W. Bush",
"Bush family"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3951,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_803
|
Wentworth Earl Miller III was the second American-British to star in a role as Michael Scofield in the Fox series Prison Break.
|
{
"evidence": [
"Wentworth Miller",
"Stoker (film)"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nWentworth Earl Miller III was the second American-British to star in a role as Michael Scofield in the Fox series Prison Break.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Wentworth Miller",
"Stoker (film)"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3952,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
64907
|
NADPH dehydrogenase is an enzyme that is involved in the chemical reaction and has two substrates.
|
{
"evidence": [
"NADPH dehydrogenase_sentence_1"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nNADPH dehydrogenase is an enzyme that is involved in the chemical reaction and has two substrates.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"NADPH dehydrogenase_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3953,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.328",
"value": "?search="
},
{
"operation": "search",
"time": "19.541",
"value": "NADPH dehydrogenase"
},
{
"operation": "Now on",
"time": "20.688",
"value": "NADPH dehydrogenase"
},
{
"operation": "Highlighting",
"time": "30.124",
"value": "NADPH dehydrogenase_sentence_0"
},
{
"operation": "Highlighting",
"time": "54.859",
"value": "NADPH dehydrogenase_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "57.339",
"value": "NADPH dehydrogenase_sentence_0"
},
{
"operation": "finish",
"time": "98.481",
"value": "finish"
}
]
}
|
29885
|
Château Climens, a wine characteristically known as better when young, though in its greatest years at the height of finesse, also produces a second wine (a second label wine made from cuvee) named Cypres de Climens.
|
{
"evidence": [
"Château Climens_sentence_3",
"Château Climens_sentence_5",
"Second wine_sentence_0"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nChâteau Climens, a wine characteristically known as better when young, though in its greatest years at the height of finesse, also produces a second wine (a second label wine made from cuvee) named Cypres de Climens.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Château Climens_sentence_3",
"Château Climens_sentence_5",
"Second wine_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3954,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.057",
"value": "?search="
},
{
"operation": "search",
"time": "16.73",
"value": "Chateau Climens"
},
{
"operation": "Now on",
"time": "17.258",
"value": "Château Climens"
},
{
"operation": "Highlighting",
"time": "36.721",
"value": "Château Climens_sentence_3"
},
{
"operation": "Highlighting",
"time": "54.771",
"value": "Château Climens_sentence_5"
},
{
"operation": "hyperlink",
"time": "85.322",
"value": "second wine"
},
{
"operation": "Now on",
"time": "86.125",
"value": "Second wine"
},
{
"operation": "Highlighting",
"time": "90.062",
"value": "Second wine_sentence_0"
},
{
"operation": "finish",
"time": "131.904",
"value": "finish"
}
]
}
|
exfever_6182
|
Henry John Deutschendorf Jr. was an American singer-songwriter who began his music career with music groups whose genre included traditional music and the contemporary genre that evolved from the former during the 20th-century folk revival.
|
{
"evidence": [
"John Denver",
"Folk music"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nHenry John Deutschendorf Jr. was an American singer-songwriter who began his music career with music groups whose genre included traditional music and the contemporary genre that evolved from the former during the 20th-century folk revival.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"John Denver",
"Folk music"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3955,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
17623
|
Pahuja moved from India to Canada with her family when she was a child, thus growing up she faced adversity as she was bombarded with a new "western lifestyle".
|
{
"evidence": [
"Nisha Pahuja_sentence_4",
"Nisha Pahuja_sentence_5"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nPahuja moved from India to Canada with her family when she was a child, thus growing up she faced adversity as she was bombarded with a new \"western lifestyle\".\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Nisha Pahuja_sentence_4",
"Nisha Pahuja_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3956,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.787",
"value": "undefined"
},
{
"operation": "search",
"time": "18.001",
"value": "Contains...Pahuja"
},
{
"operation": "Now on",
"time": "19.156",
"value": "undefined"
},
{
"operation": "search",
"time": "32.081",
"value": "\"moved from India to Canada with her family \""
},
{
"operation": "Now on",
"time": "36.507",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "38.253",
"value": "Nisha Pahuja"
},
{
"operation": "Now on",
"time": "39.598",
"value": "Nisha Pahuja"
},
{
"operation": "Page search",
"time": "44.225",
"value": "moved"
},
{
"operation": "Highlighting",
"time": "46.193",
"value": "Nisha Pahuja_sentence_4"
},
{
"operation": "Highlighting",
"time": "51.567",
"value": "Nisha Pahuja_sentence_5"
},
{
"operation": "finish",
"time": "68.422",
"value": "finish"
}
]
}
|
exfever_2383
|
The 35th president of the United States proposed the establishment of the Environmental Protection Agency, an independent executive agency of the United States federal government, on July 9, 1970.
|
{
"evidence": [
"United States Environmental Protection Agency",
"Richard Nixon"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe 35th president of the United States proposed the establishment of the Environmental Protection Agency, an independent executive agency of the United States federal government, on July 9, 1970.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"United States Environmental Protection Agency",
"Richard Nixon"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3957,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
18703
|
In 2001 and 2015 the Galagedara Polling Division claimed that UNP would win which accurately matched the country's results, as well as the Kandy electoral district results, which is one of 22 electoral districts of Sri Lanka created by the 1978 Constitution.
|
{
"evidence": [
"Galagedara Polling Division_sentence_0"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nIn 2001 and 2015 the Galagedara Polling Division claimed that UNP would win which accurately matched the country's results, as well as the Kandy electoral district results, which is one of 22 electoral districts of Sri Lanka created by the 1978 Constitution.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Galagedara Polling Division_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3958,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.466",
"value": "?search="
},
{
"operation": "search",
"time": "56.832",
"value": "2001 and 2015 the Galagedara Polling Division"
},
{
"operation": "Now on",
"time": "58.941",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "63.847",
"value": "Galagedara Polling Division"
},
{
"operation": "Now on",
"time": "65.641",
"value": "Galagedara Polling Division"
},
{
"operation": "Highlighting",
"time": "129.932",
"value": "Galagedara Polling Division_sentence_0"
},
{
"operation": "Page search",
"time": "134.88",
"value": "unp"
},
{
"operation": "Highlighting",
"time": "164.639",
"value": "Galagedara Polling Division_sentence_3"
},
{
"operation": "Highlighting",
"time": "164.643",
"value": "Galagedara Polling Division_section_10"
},
{
"operation": "Highlighting",
"time": "164.645",
"value": "Galagedara Polling Division_section_11"
},
{
"operation": "Highlighting deleted",
"time": "189.279",
"value": "Galagedara Polling Division_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "189.286",
"value": "Galagedara Polling Division_section_10"
},
{
"operation": "Highlighting deleted",
"time": "189.291",
"value": "Galagedara Polling Division_section_11"
},
{
"operation": "Highlighting",
"time": "240.638",
"value": "Galagedara Polling Division_sentence_3"
},
{
"operation": "Highlighting",
"time": "240.642",
"value": "Galagedara Polling Division_section_10"
},
{
"operation": "Highlighting",
"time": "240.643",
"value": "Galagedara Polling Division_section_11"
},
{
"operation": "Highlighting deleted",
"time": "244.738",
"value": "Galagedara Polling Division_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "244.744",
"value": "Galagedara Polling Division_section_10"
},
{
"operation": "Highlighting deleted",
"time": "244.748",
"value": "Galagedara Polling Division_section_11"
},
{
"operation": "finish",
"time": "254.339",
"value": "finish"
}
]
}
|
84400
|
Macky Ali(1970-31 August 2002 due to cardiac arrest) was an Indian film actor with polio who made his debut as a child artist in the 1974 film Ek Baap Chhe Bete and also starred in Tirchhi Topiwale and Yaro Sab Dua Karo in 1998.
|
{
"evidence": [
"Macky Ali_sentence_5"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nMacky Ali(1970-31 August 2002 due to cardiac arrest) was an Indian film actor with polio who made his debut as a child artist in the 1974 film Ek Baap Chhe Bete and also starred in Tirchhi Topiwale and Yaro Sab Dua Karo in 1998.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Macky Ali_sentence_5"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3959,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.691",
"value": "?search="
},
{
"operation": "search",
"time": "12.724",
"value": "Macky Ali"
},
{
"operation": "Now on",
"time": "13.511",
"value": "Macky Ali"
},
{
"operation": "Highlighting",
"time": "18.656",
"value": "Macky Ali_cell_0_2_1"
},
{
"operation": "Highlighting",
"time": "18.659",
"value": "Macky Ali_header_cell_0_2_0"
},
{
"operation": "Highlighting",
"time": "18.659",
"value": "Macky Ali_header_cell_0_0_0"
},
{
"operation": "Page search",
"time": "26.928",
"value": "arrest"
},
{
"operation": "Highlighting",
"time": "35.057",
"value": "Macky Ali_sentence_14"
},
{
"operation": "Highlighting deleted",
"time": "38.656",
"value": "Macky Ali_cell_0_2_1"
},
{
"operation": "Highlighting deleted",
"time": "38.66",
"value": "Macky Ali_header_cell_0_2_0"
},
{
"operation": "Highlighting deleted",
"time": "38.661",
"value": "Macky Ali_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "42.321",
"value": "Macky Ali_sentence_0"
},
{
"operation": "Page search",
"time": "57.125",
"value": "polio"
},
{
"operation": "Highlighting",
"time": "72.159",
"value": "Macky Ali_sentence_5"
},
{
"operation": "Page search",
"time": "114.592",
"value": "polio"
},
{
"operation": "Highlighting",
"time": "123.312",
"value": "Macky Ali_sentence_6"
},
{
"operation": "Highlighting deleted",
"time": "139.629",
"value": "Macky Ali_sentence_6"
},
{
"operation": "Highlighting deleted",
"time": "148.448",
"value": "Macky Ali_sentence_14"
},
{
"operation": "Highlighting deleted",
"time": "152.993",
"value": "Macky Ali_sentence_0"
},
{
"operation": "finish",
"time": "308.479",
"value": "finish"
}
]
}
|
exfever_2013
|
Ghostbusters, also known as "Ghostbusters: Answer the Call," is a 2016 American supernatural comedy film directed by an American actor, comedian, and filmmaker known for his movies such as Bridesmaids and Spy.
|
{
"evidence": [
"Ghostbusters (2016 film)",
"Paul Feig"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nGhostbusters, also known as \"Ghostbusters: Answer the Call,\" is a 2016 American supernatural comedy film directed by an American actor, comedian, and filmmaker known for his movies such as Bridesmaids and Spy.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Ghostbusters (2016 film)",
"Paul Feig"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3960,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
24227
|
John Daniel Tinder (born on February 17, 1950) was nominated to a judgeship on the United States District Court by President Ronald Reagan on June 2, 1987; and after 20 years of service as a trial judge, Tinder was nominated on July 17, 2007 by President George W. Bush to replace Judge Daniel Anthony Manion on the Seventh Circuit bench.
|
{
"evidence": [
"John Daniel Tinder_sentence_0",
"John Daniel Tinder_sentence_9",
"John Daniel Tinder_sentence_10"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJohn Daniel Tinder (born on February 17, 1950) was nominated to a judgeship on the United States District Court by President Ronald Reagan on June 2, 1987; and after 20 years of service as a trial judge, Tinder was nominated on July 17, 2007 by President George W. Bush to replace Judge Daniel Anthony Manion on the Seventh Circuit bench.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"John Daniel Tinder_sentence_0",
"John Daniel Tinder_sentence_9",
"John Daniel Tinder_sentence_10"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3961,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.608",
"value": "?search="
},
{
"operation": "search",
"time": "9.884",
"value": "John Daniel Tinder"
},
{
"operation": "Now on",
"time": "10.773",
"value": "John Daniel Tinder"
},
{
"operation": "Highlighting",
"time": "18.699",
"value": "John Daniel Tinder_sentence_0"
},
{
"operation": "Highlighting",
"time": "41.288",
"value": "John Daniel Tinder_sentence_9"
},
{
"operation": "Highlighting",
"time": "50.826",
"value": "John Daniel Tinder_sentence_10"
},
{
"operation": "finish",
"time": "75.947",
"value": "finish"
}
]
}
|
exfever_1040
|
The Conjuring is a 2016 American supernatural horror film directed by an Australian director, producer, screenwriter and comic book writer who primarily worked in the genre of speculative fiction which is intended to frighten, scare, or disgust.
|
{
"evidence": [
"The Conjuring",
"James Wan",
"Horror fiction"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Conjuring is a 2016 American supernatural horror film directed by an Australian director, producer, screenwriter and comic book writer who primarily worked in the genre of speculative fiction which is intended to frighten, scare, or disgust.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Conjuring",
"James Wan",
"Horror fiction"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3962,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8283
|
James "Clubber" Lang is a fictional character portrayed by American actor Mr. T.Lang who is a professional boxer fighting out of Chicago and a one-time world heavyweight champion, having taken the title from a fictional title character and the protagonist of the Rocky film series.
|
{
"evidence": [
"Clubber Lang",
"Rocky Balboa"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJames \"Clubber\" Lang is a fictional character portrayed by American actor Mr. T.Lang who is a professional boxer fighting out of Chicago and a one-time world heavyweight champion, having taken the title from a fictional title character and the protagonist of the Rocky film series.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Clubber Lang",
"Rocky Balboa"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3963,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4969
|
Braveheart is a historical drama film with large-scale, sweeping scope and spectacle.
|
{
"evidence": [
"Braveheart",
"Epic film"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBraveheart is a historical drama film with large-scale, sweeping scope and spectacle.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Braveheart",
"Epic film"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3964,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7303
|
George Edward Foreman is an American former professional boxer who is also an Olympic gold medalist who retired after a loss to an American heavyweight professional boxer.
|
{
"evidence": [
"George Foreman",
"Jimmy Young (boxer)"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nGeorge Edward Foreman is an American former professional boxer who is also an Olympic gold medalist who retired after a loss to an American heavyweight professional boxer.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"George Foreman",
"Jimmy Young (boxer)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3965,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
53700
|
In 2012, Algeria national football had joined in eight competitions that resulted in six winnings and two losses, and the competition they had the most winnings was in the African Cup qualification, which is the qualification process for the 2013 Africa Cup of Nations, also known as the 29th edition of the Africa Cup of Nations tournament.
|
{
"evidence": [
"2013 Africa Cup of Nations qualification_sentence_0",
"Algeria national football team_sentence_94",
"Algeria national football team_sentence_95",
"Algeria national football team_sentence_96",
"Algeria national football team_sentence_97"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nIn 2012, Algeria national football had joined in eight competitions that resulted in six winnings and two losses, and the competition they had the most winnings was in the African Cup qualification, which is the qualification process for the 2013 Africa Cup of Nations, also known as the 29th edition of the Africa Cup of Nations tournament.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"2013 Africa Cup of Nations qualification_sentence_0",
"Algeria national football team_sentence_94",
"Algeria national football team_sentence_95",
"Algeria national football team_sentence_96",
"Algeria national football team_sentence_97"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3966,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.959",
"value": "?search="
},
{
"operation": "search",
"time": "52.22",
"value": "2012 algeria national football"
},
{
"operation": "Now on",
"time": "54.016",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "57.873",
"value": "Algeria national football team results (2010–19)"
},
{
"operation": "Now on",
"time": "58.988",
"value": "Algeria national football team results (2010–19)"
},
{
"operation": "hyperlink",
"time": "62.521",
"value": "3 2012 Algeria national football team results (2010–19)_section_2"
},
{
"operation": "Highlighting",
"time": "71.521",
"value": "Algeria national football team results (2010–19)_sentence_0"
},
{
"operation": "Highlighting",
"time": "71.522",
"value": "Algeria national football team results (2010–19)_title"
},
{
"operation": "hyperlink",
"time": "72.715",
"value": "Algeria national football team"
},
{
"operation": "Now on",
"time": "75.374",
"value": "Algeria national football team"
},
{
"operation": "back-button-clicked",
"time": "104.412",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "104.514",
"value": "Algeria national football team results (2010–19)#2012 Algeria national football team results .282010.E2.80.9319.29 section 2"
},
{
"operation": "Highlighting deleted",
"time": "111.036",
"value": "Algeria national football team results (2010–19)_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "111.04",
"value": "Algeria national football team results (2010–19)_title"
},
{
"operation": "search",
"time": "144.024",
"value": "2012 Algerian Cup Final"
},
{
"operation": "Now on",
"time": "145.138",
"value": "2012 Algerian Cup Final"
},
{
"operation": "search",
"time": "172.234",
"value": "2012 African Cup of Nations qualification"
},
{
"operation": "Now on",
"time": "181.59",
"value": "2012 African Cup of Nations qualification"
},
{
"operation": "search",
"time": "218.113",
"value": "2013 Africa Cup of Nations qualification"
},
{
"operation": "Now on",
"time": "227.937",
"value": "2013 Africa Cup of Nations qualification"
},
{
"operation": "Highlighting",
"time": "247.186",
"value": "2013 Africa Cup of Nations qualification_sentence_0"
},
{
"operation": "hyperlink",
"time": "283.833",
"value": "Algeria"
},
{
"operation": "Now on",
"time": "284.424",
"value": "Algeria national football team"
},
{
"operation": "hyperlink",
"time": "312.144",
"value": "1.5.5 2011–12"
},
{
"operation": "Highlighting",
"time": "423.536",
"value": "Algeria national football team_sentence_94"
},
{
"operation": "Highlighting",
"time": "441.62",
"value": "Algeria national football team_sentence_95"
},
{
"operation": "Highlighting",
"time": "459.465",
"value": "Algeria national football team_sentence_96"
},
{
"operation": "Highlighting",
"time": "467.469",
"value": "Algeria national football team_sentence_97"
},
{
"operation": "finish",
"time": "493.544",
"value": "finish"
}
]
}
|
5004
|
Bad Press, founded in 2007, is a publisher of poetry, writings and essays, based in Toronto, and is not to be confused with another publisher of poetry by the same name, based in Manchester.
|
{
"evidence": [
"Bad Press_sentence_1"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBad Press, founded in 2007, is a publisher of poetry, writings and essays, based in Toronto, and is not to be confused with another publisher of poetry by the same name, based in Manchester.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Bad Press_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3967,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "search",
"time": "7.701",
"value": "Bad Press"
},
{
"operation": "Highlighting",
"time": "10.806",
"value": "Bad Press_sentence_1"
},
{
"operation": "finish",
"time": "22.74",
"value": "finish"
}
]
}
|
exfever_5158
|
The Boys from Brazil is a 1978 science fiction thriller film directed by Franklin J. Schaffner that stars an English actor, best known for his role as Coleman in a 1983 American comedy film directed by John Landis.
|
{
"evidence": [
"The Boys from Brazil (film)",
"Denholm Elliott",
"Trading Places"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Boys from Brazil is a 1978 science fiction thriller film directed by Franklin J. Schaffner that stars an English actor, best known for his role as Coleman in a 1983 American comedy film directed by John Landis.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Boys from Brazil (film)",
"Denholm Elliott",
"Trading Places"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3968,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
78650
|
Gloria Hatrick McLean was married twice and had two sons and two daughters.
|
{
"evidence": [
"Gloria Hatrick McLean_sentence_4",
"Gloria Hatrick McLean_sentence_8",
"Gloria Hatrick McLean_sentence_6",
"Gloria Hatrick McLean_sentence_9"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nGloria Hatrick McLean was married twice and had two sons and two daughters.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Gloria Hatrick McLean_sentence_4",
"Gloria Hatrick McLean_sentence_8",
"Gloria Hatrick McLean_sentence_6",
"Gloria Hatrick McLean_sentence_9"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3969,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.302",
"value": "?search="
},
{
"operation": "search",
"time": "52.21",
"value": "Gloria Hatrick McLean"
},
{
"operation": "Now on",
"time": "53.259",
"value": "Gloria Hatrick McLean"
},
{
"operation": "Highlighting",
"time": "70.456",
"value": "Gloria Hatrick McLean_sentence_4"
},
{
"operation": "Highlighting",
"time": "71.972",
"value": "Gloria Hatrick McLean_sentence_8"
},
{
"operation": "Highlighting",
"time": "80.848",
"value": "Gloria Hatrick McLean_sentence_6"
},
{
"operation": "Highlighting",
"time": "81.985",
"value": "Gloria Hatrick McLean_sentence_9"
},
{
"operation": "finish",
"time": "95.134",
"value": "finish"
}
]
}
|
22667
|
The Jericho Union Free School District operates 18 schools and employs 564 teachers according to the 2002-2003 school year data.
|
{
"evidence": [
"Jericho Union Free School District_sentence_2"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Jericho Union Free School District operates 18 schools and employs 564 teachers according to the 2002-2003 school year data.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Jericho Union Free School District_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3970,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.367",
"value": "?search="
},
{
"operation": "search",
"time": "34.084",
"value": "Jericho Union Free School District"
},
{
"operation": "Now on",
"time": "35.224",
"value": "Jericho Union Free School District"
},
{
"operation": "Highlighting",
"time": "44.176",
"value": "Jericho Union Free School District_sentence_2"
},
{
"operation": "finish",
"time": "95.917",
"value": "finish"
}
]
}
|
93899
|
Linfen is located in the southwestern part of Shanxi with the Yellow River to the west and it spans 170 kilometers.
|
{
"evidence": [
"Linfen_sentence_0",
"Linfen_sentence_21",
"Linfen_sentence_20"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nLinfen is located in the southwestern part of Shanxi with the Yellow River to the west and it spans 170 kilometers.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Linfen_sentence_0",
"Linfen_sentence_21",
"Linfen_sentence_20"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3971,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.04",
"value": "?search="
},
{
"operation": "search",
"time": "14.442",
"value": "Linfen"
},
{
"operation": "Now on",
"time": "15.783",
"value": "Linfen"
},
{
"operation": "Highlighting",
"time": "46.756",
"value": "Linfen_sentence_0"
},
{
"operation": "Highlighting",
"time": "66.536",
"value": "Linfen_sentence_21"
},
{
"operation": "Highlighting",
"time": "67.508",
"value": "Linfen_sentence_20"
},
{
"operation": "finish",
"time": "140.057",
"value": "finish"
}
]
}
|
exfever_1819
|
James Rash is the 21st American actor who has received a Golden Globe nomination.
|
{
"evidence": [
"Jim Rash",
"NBC"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJames Rash is the 21st American actor who has received a Golden Globe nomination.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Jim Rash",
"NBC"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3972,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5775
|
Robert Siers Cavallo was the biggest-selling producer in history and the first recipient of the Grammy award.
|
{
"evidence": [
"Rob Cavallo",
"Kid Rock"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nRobert Siers Cavallo was the biggest-selling producer in history and the first recipient of the Grammy award.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Rob Cavallo",
"Kid Rock"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3973,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
58068
|
A rug plot is used to visualize the distribution of the data and in combination with three-dimensional scatter plots by placing a rug plot of the x values of the data along the x-axis, and similarly for the y values.
|
{
"evidence": [
"Rug plot_sentence_4"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nA rug plot is used to visualize the distribution of the data and in combination with three-dimensional scatter plots by placing a rug plot of the x values of the data along the x-axis, and similarly for the y values.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Rug plot_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3974,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.156",
"value": "?search="
},
{
"operation": "search",
"time": "11.388",
"value": "Rug plot"
},
{
"operation": "Now on",
"time": "12.183",
"value": "Rug plot"
},
{
"operation": "Highlighting",
"time": "24.008",
"value": "Rug plot_sentence_2"
},
{
"operation": "Highlighting",
"time": "38.95",
"value": "Rug plot_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "45.622",
"value": "Rug plot_sentence_2"
},
{
"operation": "Page search",
"time": "66.557",
"value": "three-dimensional"
},
{
"operation": "finish",
"time": "88.513",
"value": "finish"
}
]
}
|
exfever_7740
|
Inside Man is a 2006 heist thriller directed by an American film director and producer.
|
{
"evidence": [
"Inside Man",
"Spike Lee"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nInside Man is a 2006 heist thriller directed by an American film director and producer.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Inside Man",
"Spike Lee"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3975,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
29543
|
The entrance to Westville Dam is on Wallace Road in Sturbridge, and camping and swimming are encouraged there.
|
{
"evidence": [
"Westville Dam_sentence_15"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe entrance to Westville Dam is on Wallace Road in Sturbridge, and camping and swimming are encouraged there.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Westville Dam_sentence_15"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3976,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.657",
"value": "?search="
},
{
"operation": "search",
"time": "8.411",
"value": "Westville Dam"
},
{
"operation": "Now on",
"time": "8.866",
"value": "Westville Dam"
},
{
"operation": "Highlighting",
"time": "19.554",
"value": "Westville Dam_sentence_13"
},
{
"operation": "Highlighting",
"time": "26.082",
"value": "Westville Dam_sentence_15"
},
{
"operation": "Highlighting deleted",
"time": "27.46",
"value": "Westville Dam_sentence_13"
},
{
"operation": "finish",
"time": "32.859",
"value": "finish"
}
]
}
|
21617
|
Kohzoh Imai received her MD from Osaka University in 1972 and Ph.D. five years later.
|
{
"evidence": [
"Kohzoh Imai_sentence_11",
"Kohzoh Imai_sentence_12"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nKohzoh Imai received her MD from Osaka University in 1972 and Ph.D. five years later.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Kohzoh Imai_sentence_11",
"Kohzoh Imai_sentence_12"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3977,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.651",
"value": "?search="
},
{
"operation": "search",
"time": "8.974",
"value": "Kohzoh Imai"
},
{
"operation": "Now on",
"time": "9.835",
"value": "Kohzoh Imai"
},
{
"operation": "Highlighting",
"time": "21.424",
"value": "Kohzoh Imai_sentence_11"
},
{
"operation": "Highlighting",
"time": "39.044",
"value": "Kohzoh Imai_sentence_12"
},
{
"operation": "Highlighting",
"time": "47.488",
"value": "Kohzoh Imai_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "49.448",
"value": "Kohzoh Imai_sentence_3"
},
{
"operation": "search",
"time": "52.673",
"value": "Sapporo Medical University"
},
{
"operation": "Now on",
"time": "53.765",
"value": "Sapporo Medical University"
},
{
"operation": "back-button-clicked",
"time": "62.281",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "62.365",
"value": "Kohzoh Imai"
},
{
"operation": "finish",
"time": "68.789",
"value": "finish"
}
]
}
|
exfever_1119
|
Sir Richard Starkey is an English musician who occasionally sang lead vocals with the group, usually for one song on each album, including a song by the English rock band the Beatles from their seventh studio album that topped the Record Retailer chart in the UK for seven weeks and the US Billboard Top LPs list for six weeks.
|
{
"evidence": [
"Ringo Starr",
"Yellow Submarine (song)",
"Revolver (Beatles album)"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nSir Richard Starkey is an English musician who occasionally sang lead vocals with the group, usually for one song on each album, including a song by the English rock band the Beatles from their seventh studio album that topped the Record Retailer chart in the UK for seven weeks and the US Billboard Top LPs list for six weeks.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Ringo Starr",
"Yellow Submarine (song)",
"Revolver (Beatles album)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3978,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
54728
|
Hang is believed to derive from the Old English word Hangar, the place is mentioned in the Domesday Book as Hotun and they were divided up into nine parishes.
|
{
"evidence": [
"Hang East_sentence_3",
"Hang East_sentence_4",
"Hang East_sentence_6"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nHang is believed to derive from the Old English word Hangar, the place is mentioned in the Domesday Book as Hotun and they were divided up into nine parishes.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Hang East_sentence_3",
"Hang East_sentence_4",
"Hang East_sentence_6"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3979,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.9",
"value": "?search="
},
{
"operation": "search",
"time": "6.366",
"value": "Hang"
},
{
"operation": "Now on",
"time": "7.497",
"value": "Hang"
},
{
"operation": "Page search",
"time": "18.049",
"value": "hangar"
},
{
"operation": "search",
"time": "37.673",
"value": "\"Hang is believed to derive from the Old English word Hangar\""
},
{
"operation": "Now on",
"time": "38.676",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "41.202",
"value": "Hang East"
},
{
"operation": "Now on",
"time": "42.379",
"value": "Hang East"
},
{
"operation": "Page search",
"time": "107.94",
"value": "Hangar"
},
{
"operation": "Highlighting",
"time": "109.588",
"value": "Hang East_sentence_3"
},
{
"operation": "Highlighting",
"time": "123.36",
"value": "Hang East_sentence_4"
},
{
"operation": "Page search",
"time": "127.72",
"value": "nine"
},
{
"operation": "Page search",
"time": "129.911",
"value": "nine parish"
},
{
"operation": "Highlighting",
"time": "131.32",
"value": "Hang East_sentence_6"
},
{
"operation": "finish",
"time": "165.725",
"value": "finish"
}
]
}
|
72328
|
The 1964 Wandsworth London Borough Council election had a total of 169 candidates inclusive of Conservative, Labour, Liberal, Communists, Independent, and Rad Lib. parties wherein the election was only based on aldermen.
|
{
"evidence": [
"1964 Wandsworth London Borough Council election_sentence_8"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe 1964 Wandsworth London Borough Council election had a total of 169 candidates inclusive of Conservative, Labour, Liberal, Communists, Independent, and Rad Lib. parties wherein the election was only based on aldermen.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"1964 Wandsworth London Borough Council election_sentence_8"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3980,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.833",
"value": "?search="
},
{
"operation": "search",
"time": "10.269",
"value": "1964 Wandsworth London Borough Council election"
},
{
"operation": "Now on",
"time": "11.323",
"value": "1964 Wandsworth London Borough Council election"
},
{
"operation": "Highlighting",
"time": "14.7",
"value": "1964 Wandsworth London Borough Council election_sentence_5"
},
{
"operation": "Highlighting",
"time": "18.147",
"value": "1964 Wandsworth London Borough Council election_sentence_6"
},
{
"operation": "Highlighting",
"time": "24.303",
"value": "1964 Wandsworth London Borough Council election_sentence_7"
},
{
"operation": "Highlighting",
"time": "35.21",
"value": "1964 Wandsworth London Borough Council election_sentence_8"
},
{
"operation": "Highlighting deleted",
"time": "40.75",
"value": "1964 Wandsworth London Borough Council election_sentence_5"
},
{
"operation": "Highlighting deleted",
"time": "41.197",
"value": "1964 Wandsworth London Borough Council election_sentence_6"
},
{
"operation": "Highlighting deleted",
"time": "41.624",
"value": "1964 Wandsworth London Borough Council election_sentence_7"
},
{
"operation": "finish",
"time": "47.51",
"value": "finish"
}
]
}
|
16399
|
In San Antonio Riders season 1991 l (American football team), WLAF league got six wins and four lost while in 1992 got three wins and seven lost.
|
{
"evidence": [
"San Antonio Riders_sentence_0"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nIn San Antonio Riders season 1991 l (American football team), WLAF league got six wins and four lost while in 1992 got three wins and seven lost.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"San Antonio Riders_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3981,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.367",
"value": "?search="
},
{
"operation": "search",
"time": "8.025",
"value": "San Antonio Riders season 1991"
},
{
"operation": "Now on",
"time": "10.117",
"value": "undefined"
},
{
"operation": "search",
"time": "53.954",
"value": "San Antonio Riders season 1991 l"
},
{
"operation": "Now on",
"time": "55.301",
"value": "undefined"
},
{
"operation": "search",
"time": "91.363",
"value": "San Antonio Riders 1991"
},
{
"operation": "Now on",
"time": "92.316",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "95.961",
"value": "Week 6"
},
{
"operation": "Now on",
"time": "97.254",
"value": "1991 Birmingham Fire season#Week 6"
},
{
"operation": "hyperlink",
"time": "99.292",
"value": "San Antonio Riders"
},
{
"operation": "Now on",
"time": "100.446",
"value": "San Antonio Riders"
},
{
"operation": "Highlighting",
"time": "101.586",
"value": "San Antonio Riders_sentence_0"
},
{
"operation": "back-button-clicked",
"time": "104.63",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "104.822",
"value": "1991 Birmingham Fire season#Week 6"
},
{
"operation": "search",
"time": "262.753",
"value": "WLAF 1991"
},
{
"operation": "Now on",
"time": "264.494",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "272.357",
"value": "1991 Birmingham Fire season"
},
{
"operation": "Now on",
"time": "273.281",
"value": "1991 Birmingham Fire season"
},
{
"operation": "Page search",
"time": "291.16",
"value": "1992"
},
{
"operation": "search",
"time": "313.572",
"value": "WLAF league got six wins and four lost while in 1992 got three wins and seven lost"
},
{
"operation": "Now on",
"time": "315.404",
"value": "undefined"
},
{
"operation": "finish",
"time": "332.355",
"value": "finish"
}
]
}
|
37913
|
The three Spanish athletes who competed at the 1954 European Athletics Championships in Bern, Switzerland, from 25–29 August 1954, did not advance to the semi-finals.
|
{
"evidence": [
"1954 European Athletics Championships_sentence_0"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe three Spanish athletes who competed at the 1954 European Athletics Championships in Bern, Switzerland, from 25–29 August 1954, did not advance to the semi-finals.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"1954 European Athletics Championships_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3982,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.657",
"value": "?search="
},
{
"operation": "search",
"time": "34.217",
"value": "1954 European Athletics Championships"
},
{
"operation": "Now on",
"time": "35.425",
"value": "1954 European Athletics Championships"
},
{
"operation": "Highlighting",
"time": "40.585",
"value": "1954 European Athletics Championships_sentence_0"
},
{
"operation": "search",
"time": "160.828",
"value": "1954 european athletics championship spain"
},
{
"operation": "Now on",
"time": "162.872",
"value": "undefined"
},
{
"operation": "search",
"time": "330.137",
"value": "he three Spanish athletes who competed at the 1954 European Athletics Championships in Bern, Switzerland, from 25–29 August 1954, did not advance to the semi-finals"
},
{
"operation": "Now on",
"time": "332.338",
"value": "undefined"
},
{
"operation": "search",
"time": "357.195",
"value": "spanish athletes 1954 european athletics"
},
{
"operation": "Now on",
"time": "358.018",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "371.924",
"value": "List of European Athletics Championships records"
},
{
"operation": "Now on",
"time": "373.043",
"value": "List of European Athletics Championships records"
},
{
"operation": "search",
"time": "398.235",
"value": "1954 European Athletics Championships – Men's 10,000 metres track walk"
},
{
"operation": "Now on",
"time": "399.138",
"value": "1954 European Athletics Championships – Men's 10,000 metres track walk"
},
{
"operation": "search",
"time": "434.308",
"value": "1954 european athletics championships semi finals"
},
{
"operation": "Now on",
"time": "435.437",
"value": "undefined"
},
{
"operation": "search",
"time": "454.939",
"value": "1954 European Athletics Championships"
},
{
"operation": "Now on",
"time": "455.678",
"value": "1954 European Athletics Championships"
},
{
"operation": "hyperlink",
"time": "474.217",
"value": "Soviet Union"
},
{
"operation": "Now on",
"time": "475.308",
"value": "Soviet Union"
},
{
"operation": "back-button-clicked",
"time": "477.236",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "478.712",
"value": "1954 European Athletics Championships"
},
{
"operation": "finish",
"time": "509.793",
"value": "finish"
}
]
}
|
38407
|
According to the IEA, a policy advisor focusing on energy security, economic development, and environmental protection, energy security and renewable technology could make a direct contribution to heating.
|
{
"evidence": [
"International Energy Agency_sentence_2",
"International Energy Agency_sentence_3",
"International Energy Agency_sentence_5"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nAccording to the IEA, a policy advisor focusing on energy security, economic development, and environmental protection, energy security and renewable technology could make a direct contribution to heating.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"International Energy Agency_sentence_2",
"International Energy Agency_sentence_3",
"International Energy Agency_sentence_5"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3983,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.952",
"value": "?search="
},
{
"operation": "search",
"time": "30.782",
"value": "IEA"
},
{
"operation": "Now on",
"time": "31.45",
"value": "IEA"
},
{
"operation": "hyperlink",
"time": "51.829",
"value": "International Energy Agency"
},
{
"operation": "Now on",
"time": "52.612",
"value": "International Energy Agency"
},
{
"operation": "Highlighting",
"time": "56.868",
"value": "International Energy Agency_sentence_2"
},
{
"operation": "Highlighting",
"time": "63.349",
"value": "International Energy Agency_sentence_3"
},
{
"operation": "Highlighting",
"time": "76.243",
"value": "International Energy Agency_sentence_5"
},
{
"operation": "back-button-clicked",
"time": "135.691",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "135.907",
"value": "IEA"
},
{
"operation": "hyperlink",
"time": "153.87",
"value": "International Energy Agency"
},
{
"operation": "Now on",
"time": "154.398",
"value": "International Energy Agency"
},
{
"operation": "search",
"time": "189.899",
"value": "renewable technology"
},
{
"operation": "Now on",
"time": "191.678",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "196.459",
"value": "Renewable energy"
},
{
"operation": "Now on",
"time": "198.284",
"value": "Renewable energy"
},
{
"operation": "finish",
"time": "210.014",
"value": "finish"
}
]
}
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.