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 |
|---|---|---|---|---|---|---|---|---|
exfever_6062
|
The Apple Store is a chain of retail stores owned and operated by Intel that sell various Apple products, including a collection of instructions that tell a computer how to work.
|
{
"evidence": [
"Apple Store",
"Software"
],
"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 Apple Store is a chain of retail stores owned and operated by Intel that sell various Apple products, including a collection of instructions that tell a computer how to work.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Apple Store",
"Software"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1684,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6410
|
Con Man is a Candian comedy web series directed by Alan Tudyk, who was also one of the stars of the 2002 Western drama television series directed by Joss Whedon.
|
{
"evidence": [
"Con Man (web series)",
"Firefly (TV series)"
],
"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:\nCon Man is a Candian comedy web series directed by Alan Tudyk, who was also one of the stars of the 2002 Western drama television series directed by Joss Whedon.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Con Man (web series)",
"Firefly (TV series)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1685,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
41082
|
The dorsal sepal is egg-shaped, 9–13 mm long, 4–6.5 mm wide and curves upwards; while the sepals (function as protection for the flower in bud) are oblong, 9–14 mm long, 2–3 mm wide and turned downwards.
|
{
"evidence": [
"Diuris eburnea_sentence_5",
"Diuris eburnea_sentence_6",
"Sepal_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:\nThe dorsal sepal is egg-shaped, 9–13 mm long, 4–6.5 mm wide and curves upwards; while the sepals (function as protection for the flower in bud) are oblong, 9–14 mm long, 2–3 mm wide and turned downwards.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Diuris eburnea_sentence_5",
"Diuris eburnea_sentence_6",
"Sepal_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1686,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.678",
"value": "?search="
},
{
"operation": "search",
"time": "38.891",
"value": "Contains...dorsal sepal is egg-shaped"
},
{
"operation": "Now on",
"time": "40.468",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "78.319",
"value": "Diuris eburnea"
},
{
"operation": "Now on",
"time": "79.101",
"value": "Diuris eburnea"
},
{
"operation": "Highlighting",
"time": "95.155",
"value": "Diuris eburnea_sentence_5"
},
{
"operation": "Highlighting",
"time": "107.839",
"value": "Diuris eburnea_sentence_6"
},
{
"operation": "hyperlink",
"time": "118.031",
"value": "sepal"
},
{
"operation": "Now on",
"time": "118.582",
"value": "Sepal"
},
{
"operation": "Highlighting",
"time": "128.17",
"value": "Sepal_sentence_1"
},
{
"operation": "finish",
"time": "140.9",
"value": "finish"
}
]
}
|
88143
|
Al Helfer and Rex Barney called the games for New York's WOR-TV; however, due to competition with games at Yankee Stadium, Phillies games were not able to match the audience for Yankee broadcasts.
|
{
"evidence": [
"1958 Philadelphia Phillies season_sentence_6",
"1958 Philadelphia Phillies season_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:\nAl Helfer and Rex Barney called the games for New York's WOR-TV; however, due to competition with games at Yankee Stadium, Phillies games were not able to match the audience for Yankee broadcasts.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"1958 Philadelphia Phillies season_sentence_6",
"1958 Philadelphia Phillies season_sentence_7"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1687,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.628",
"value": "?search="
},
{
"operation": "search",
"time": "12.516",
"value": "Contains...New York's WOR-TV"
},
{
"operation": "Now on",
"time": "14.574",
"value": "undefined"
},
{
"operation": "search",
"time": "34.172",
"value": "Al Helfer and Rex Barney"
},
{
"operation": "Now on",
"time": "35.285",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "37.994",
"value": "Awards and honors 1958 Philadelphia Phillies season_section_16"
},
{
"operation": "Now on",
"time": "39.644",
"value": "1958 Philadelphia Phillies season#Awards and honors 1958 Philadelphia Phillies season section 16"
},
{
"operation": "Highlighting",
"time": "364.297",
"value": "1958 Philadelphia Phillies season_sentence_6"
},
{
"operation": "Highlighting",
"time": "365.579",
"value": "1958 Philadelphia Phillies season_sentence_7"
},
{
"operation": "finish",
"time": "378.12",
"value": "finish"
}
]
}
|
35376
|
Open Bible Standard Churches can be traced to the Azusa Street Revival and the stark opposition to the teachings of Martin Luther.
|
{
"evidence": [
"Open Bible Standard Churches_sentence_8"
],
"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:\nOpen Bible Standard Churches can be traced to the Azusa Street Revival and the stark opposition to the teachings of Martin Luther.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Open Bible Standard Churches_sentence_8"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1688,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.237",
"value": "undefined"
},
{
"operation": "search",
"time": "5.622",
"value": "Open Bible Standard Churches"
},
{
"operation": "Now on",
"time": "6.599",
"value": "Open Bible Standard Churches"
},
{
"operation": "Page search",
"time": "33.182",
"value": "Azusa Street Revival"
},
{
"operation": "Highlighting",
"time": "62.249",
"value": "Open Bible Standard Churches_sentence_8"
},
{
"operation": "Page search",
"time": "113.016",
"value": "martin luther"
},
{
"operation": "Page search",
"time": "122.744",
"value": "luther"
},
{
"operation": "Page search",
"time": "123.194",
"value": "luther"
},
{
"operation": "hyperlink",
"time": "136.329",
"value": "Azusa Street Revival"
},
{
"operation": "Now on",
"time": "136.914",
"value": "Azusa Street Revival"
},
{
"operation": "Page search",
"time": "152.084",
"value": "martin luther"
},
{
"operation": "Page search",
"time": "156.734",
"value": "martin luther"
},
{
"operation": "back-button-clicked",
"time": "175.875",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "175.931",
"value": "Open Bible Standard Churches"
},
{
"operation": "hyperlink",
"time": "252.189",
"value": "Pentecostal"
},
{
"operation": "Now on",
"time": "253.679",
"value": "Pentecostalism"
},
{
"operation": "Page search",
"time": "260.782",
"value": "martin luther"
},
{
"operation": "back-button-clicked",
"time": "274.641",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "274.704",
"value": "Open Bible Standard Churches"
},
{
"operation": "search",
"time": "283.803",
"value": "martin luther"
},
{
"operation": "Now on",
"time": "284.887",
"value": "Martin Luther"
},
{
"operation": "finish",
"time": "774.067",
"value": "finish"
}
]
}
|
52339
|
Deon Kipping scheduled the release of his Billboard ranking album, I Just Want to Hear You, in September 4, 2012 under Verity Records, however, it was postponed to a later date.
|
{
"evidence": [
"Deon Kipping_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:\nDeon Kipping scheduled the release of his Billboard ranking album, I Just Want to Hear You, in September 4, 2012 under Verity Records, however, it was postponed to a later date.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Deon Kipping_sentence_9"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1689,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.669",
"value": "?search="
},
{
"operation": "search",
"time": "10.555",
"value": "Deon Kipping"
},
{
"operation": "Now on",
"time": "11.294",
"value": "Deon Kipping"
},
{
"operation": "Page search",
"time": "22.325",
"value": "September 4"
},
{
"operation": "Highlighting",
"time": "26.528",
"value": "Deon Kipping_sentence_9"
},
{
"operation": "finish",
"time": "96.359",
"value": "finish"
}
]
}
|
exfever_2948
|
Gia is a film about the life and times of Gia Carangi, one of the first highly paid fashion models who has a worldwide reputation.
|
{
"evidence": [
"Gia",
"Supermodel"
],
"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:\nGia is a film about the life and times of Gia Carangi, one of the first highly paid fashion models who has a worldwide reputation.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Gia",
"Supermodel"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1690,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
57597
|
One manual depicted Walt Kelly's friend Ward Kimball as a caveman in the same period that saw the creation of Kelly's most famous character, Pogo.
|
{
"evidence": [
"Walt Kelly_sentence_22",
"Walt Kelly_sentence_23"
],
"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:\nOne manual depicted Walt Kelly's friend Ward Kimball as a caveman in the same period that saw the creation of Kelly's most famous character, Pogo.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Walt Kelly_sentence_22",
"Walt Kelly_sentence_23"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1691,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.804",
"value": "?search="
},
{
"operation": "search",
"time": "17.511",
"value": "Ward Kimball"
},
{
"operation": "Now on",
"time": "18.654",
"value": "Ward Kimball"
},
{
"operation": "Page search",
"time": "43.506",
"value": "cavemen"
},
{
"operation": "Page search",
"time": "44.978",
"value": "caveman"
},
{
"operation": "Page search",
"time": "55.313",
"value": "cave"
},
{
"operation": "search",
"time": "62",
"value": "Walt Kelly"
},
{
"operation": "Now on",
"time": "62.657",
"value": "Walt Kelly"
},
{
"operation": "Highlighting",
"time": "64.908",
"value": "Walt Kelly_sentence_22"
},
{
"operation": "Highlighting",
"time": "68.546",
"value": "Walt Kelly_sentence_23"
},
{
"operation": "finish",
"time": "97.784",
"value": "finish"
}
]
}
|
exfever_7430
|
Michael Pence?served as a member of one of the two major contemporary political parties in the United Kingdom.
|
{
"evidence": [
"Mike Pence",
"Republican Party (United States)"
],
"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:\nMichael Pence?served as a member of one of the two major contemporary political parties in the United Kingdom.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Mike Pence",
"Republican Party (United States)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1692,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
17020
|
In Khamsa of Nizami, Layla and Majnun, is the subject of the first of his four romances where he reworked the material.
|
{
"evidence": [
"Khamsa of Nizami_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:\nIn Khamsa of Nizami, Layla and Majnun, is the subject of the first of his four romances where he reworked the material.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Khamsa of Nizami_sentence_5"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1693,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.963",
"value": "?search="
},
{
"operation": "search",
"time": "18.288",
"value": "Khamsa of Nizami"
},
{
"operation": "Now on",
"time": "19.481",
"value": "Khamsa of Nizami"
},
{
"operation": "Highlighting",
"time": "42.755",
"value": "Khamsa of Nizami_sentence_5"
},
{
"operation": "finish",
"time": "57.785",
"value": "finish"
}
]
}
|
76894
|
Nicaraguan University of Science and Technology, a private university in Managua, Nicaragua, was established along with Asia's oldest university.
|
{
"evidence": [
"Nicaraguan University of Science and Technology_sentence_0",
"History of the University of Santo Tomas_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:\nNicaraguan University of Science and Technology, a private university in Managua, Nicaragua, was established along with Asia's oldest university.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Nicaraguan University of Science and Technology_sentence_0",
"History of the University of Santo Tomas_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1694,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.315",
"value": "?search="
},
{
"operation": "Now on",
"time": "801.189",
"value": "?search="
},
{
"operation": "search",
"time": "811.604",
"value": "Nicaraguan University of Science and Technology"
},
{
"operation": "Now on",
"time": "812.71",
"value": "Nicaraguan University of Science and Technology"
},
{
"operation": "Highlighting",
"time": "848.288",
"value": "Nicaraguan University of Science and Technology_sentence_0"
},
{
"operation": "search",
"time": "870.658",
"value": "was established along with Asia's oldest university"
},
{
"operation": "Now on",
"time": "873.627",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "880.489",
"value": "Asia Manila_section_38"
},
{
"operation": "Now on",
"time": "882.061",
"value": "Manila#Asia Manila section 38"
},
{
"operation": "search",
"time": "900.759",
"value": "asia's oldest university"
},
{
"operation": "Now on",
"time": "901.865",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "909.625",
"value": "University of Santo Tomas (UST) Dispute over the oldest school in the Philippines_section_1"
},
{
"operation": "Now on",
"time": "910.059",
"value": "Dispute over the oldest school in the Philippines#University of Santo Tomas .28UST.29 Dispute over the oldest school in the Philippines section 1"
},
{
"operation": "hyperlink",
"time": "937.842",
"value": "History of the University of Santo Tomas"
},
{
"operation": "Now on",
"time": "939.131",
"value": "History of the University of Santo Tomas"
},
{
"operation": "Highlighting",
"time": "945.131",
"value": "History of the University of Santo Tomas_sentence_2"
},
{
"operation": "back-button-clicked",
"time": "947.371",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "947.468",
"value": "Dispute over the oldest school in the Philippines#University of Santo Tomas .28UST.29 Dispute over the oldest school in the Philippines section 1"
},
{
"operation": "back-button-clicked",
"time": "947.81",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "947.818",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "948.631",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "949.381",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "949.391",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "949.711",
"value": "Manila#Asia Manila section 38"
},
{
"operation": "back-button-clicked",
"time": "950.524",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "950.531",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "951.442",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "952.266",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "952.273",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "952.369",
"value": "Nicaraguan University of Science and Technology"
},
{
"operation": "Highlighting",
"time": "956.252",
"value": "Nicaraguan University of Science and Technology_cell_0_3_1"
},
{
"operation": "Highlighting",
"time": "956.261",
"value": "Nicaraguan University of Science and Technology_header_cell_0_3_0"
},
{
"operation": "Highlighting deleted",
"time": "958.784",
"value": "Nicaraguan University of Science and Technology_cell_0_3_1"
},
{
"operation": "Highlighting deleted",
"time": "958.791",
"value": "Nicaraguan University of Science and Technology_header_cell_0_3_0"
},
{
"operation": "finish",
"time": "962.44",
"value": "finish"
}
]
}
|
21555
|
Radochów has a population of 500 which has always been located in Poland where they speak Polish.
|
{
"evidence": [
"Radochów_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:\nRadochów has a population of 500 which has always been located in Poland where they speak Polish.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Radochów_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1695,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.566",
"value": "?search="
},
{
"operation": "search",
"time": "15.039",
"value": "Radochów"
},
{
"operation": "Now on",
"time": "15.692",
"value": "Radochów"
},
{
"operation": "Highlighting",
"time": "27.688",
"value": "Radochów_header_cell_0_6_0"
},
{
"operation": "Highlighting",
"time": "28.576",
"value": "Radochów_header_cell_0_7_0"
},
{
"operation": "Highlighting",
"time": "29.872",
"value": "Radochów_cell_0_7_1"
},
{
"operation": "Highlighting",
"time": "29.875",
"value": "Radochów_header_cell_0_7_0"
},
{
"operation": "Highlighting",
"time": "29.876",
"value": "Radochów_header_cell_0_6_0"
},
{
"operation": "Highlighting",
"time": "31.769",
"value": "Radochów_header_cell_0_0_0"
},
{
"operation": "Highlighting deleted",
"time": "35.554",
"value": "Radochów_header_cell_0_0_0"
},
{
"operation": "Highlighting deleted",
"time": "37.827",
"value": "Radochów_cell_0_7_1"
},
{
"operation": "Highlighting deleted",
"time": "37.834",
"value": "Radochów_header_cell_0_7_0"
},
{
"operation": "Highlighting deleted",
"time": "37.835",
"value": "Radochów_header_cell_0_6_0"
},
{
"operation": "Highlighting",
"time": "38.863",
"value": "Radochów_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "40.409",
"value": "Radochów_cell_0_7_1"
},
{
"operation": "Highlighting",
"time": "40.413",
"value": "Radochów_header_cell_0_7_0"
},
{
"operation": "Highlighting",
"time": "40.413",
"value": "Radochów_header_cell_0_6_0"
},
{
"operation": "Highlighting",
"time": "63.904",
"value": "Radochów_sentence_0"
},
{
"operation": "Highlighting",
"time": "65.431",
"value": "Radochów_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "68.439",
"value": "Radochów_cell_0_7_1"
},
{
"operation": "Highlighting deleted",
"time": "68.445",
"value": "Radochów_header_cell_0_7_0"
},
{
"operation": "Highlighting deleted",
"time": "68.446",
"value": "Radochów_header_cell_0_6_0"
},
{
"operation": "Highlighting deleted",
"time": "69.685",
"value": "Radochów_header_cell_0_0_0"
},
{
"operation": "Highlighting deleted",
"time": "104.887",
"value": "Radochów_sentence_0"
},
{
"operation": "finish",
"time": "209.877",
"value": "finish"
}
]
}
|
exfever_8678
|
Shallow Hal was released in 2001 by 20th Century Fox and grossed $141 million.
|
{
"evidence": [
"Shallow Hal",
"Jason Alexander"
],
"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:\nShallow Hal was released in 2001 by 20th Century Fox and grossed $141 million.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Shallow Hal",
"Jason Alexander"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1696,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
78295
|
Symphyllocarpus used to be classified into Kingdom Plantae and family Asteraceae but now it is classified into Kingdom Eubacteria.
|
{
"evidence": [
"Symphyllocarpus_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:\nSymphyllocarpus used to be classified into Kingdom Plantae and family Asteraceae but now it is classified into Kingdom Eubacteria.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Symphyllocarpus_sentence_1"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1697,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.754",
"value": "?search="
},
{
"operation": "search",
"time": "14.246",
"value": "Symphyllocarpus"
},
{
"operation": "Now on",
"time": "14.81",
"value": "Symphyllocarpus"
},
{
"operation": "Highlighting",
"time": "44.926",
"value": "Symphyllocarpus_cell_0_2_1"
},
{
"operation": "Highlighting",
"time": "44.928",
"value": "Symphyllocarpus_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "44.928",
"value": "Symphyllocarpus_header_cell_0_0_0"
},
{
"operation": "Highlighting deleted",
"time": "52.605",
"value": "Symphyllocarpus_cell_0_2_1"
},
{
"operation": "Highlighting deleted",
"time": "52.607",
"value": "Symphyllocarpus_header_cell_0_1_0"
},
{
"operation": "Highlighting deleted",
"time": "52.608",
"value": "Symphyllocarpus_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "65.013",
"value": "Symphyllocarpus_cell_0_2_0"
},
{
"operation": "Highlighting",
"time": "65.014",
"value": "Symphyllocarpus_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "65.014",
"value": "Symphyllocarpus_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "65.645",
"value": "Symphyllocarpus_cell_0_2_1"
},
{
"operation": "Highlighting",
"time": "78.748",
"value": "Symphyllocarpus_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "80.98",
"value": "Symphyllocarpus_cell_0_2_1"
},
{
"operation": "Highlighting deleted",
"time": "82.508",
"value": "Symphyllocarpus_cell_0_2_0"
},
{
"operation": "Highlighting deleted",
"time": "82.51",
"value": "Symphyllocarpus_header_cell_0_1_0"
},
{
"operation": "Highlighting deleted",
"time": "82.511",
"value": "Symphyllocarpus_header_cell_0_0_0"
},
{
"operation": "hyperlink",
"time": "84.204",
"value": "Symphyllocarpus exilis"
},
{
"operation": "Now on",
"time": "84.843",
"value": "Symphyllocarpus exilis"
},
{
"operation": "search",
"time": "103.078",
"value": "Eubacteria"
},
{
"operation": "Now on",
"time": "104.35",
"value": "Bacteria"
},
{
"operation": "finish",
"time": "159.077",
"value": "finish"
}
]
}
|
5893
|
Brian Pockar was a three-time national champion American figure skater.
|
{
"evidence": [
"Brian Pockar_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:\nBrian Pockar was a three-time national champion American figure skater.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Brian Pockar_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1698,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.56",
"value": "?search="
},
{
"operation": "search",
"time": "12.315",
"value": "Brian Pockar"
},
{
"operation": "Now on",
"time": "13.222",
"value": "Brian Pockar"
},
{
"operation": "Highlighting",
"time": "24.433",
"value": "Brian Pockar_sentence_0"
},
{
"operation": "finish",
"time": "33.249",
"value": "finish"
}
]
}
|
74386
|
Table tennis at the Asian Games (Table tennis has been contested at the Asian Games since 1958 except the sports were excluded from the program in 1970, with singles and doubles events for both men and women).
|
{
"evidence": [
"Table tennis at the Asian Games_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:\nTable tennis at the Asian Games (Table tennis has been contested at the Asian Games since 1958 except the sports were excluded from the program in 1970, with singles and doubles events for both men and women).\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Table tennis at the Asian Games_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1699,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.04",
"value": "?search="
},
{
"operation": "search",
"time": "52.221",
"value": "Table tennis at the Asian Games"
},
{
"operation": "Now on",
"time": "53.621",
"value": "Table tennis at the Asian Games"
},
{
"operation": "Highlighting",
"time": "84.195",
"value": "Table tennis at the Asian Games_sentence_0"
},
{
"operation": "finish",
"time": "103.071",
"value": "finish"
}
]
}
|
exfever_4757
|
Rowan Sebastian Atkinson is an Irish actor, comedian and writer who played the title role in a series of four period British sitcoms, plus several one-off installments, which originally aired on BBC One from 1983 to 1989.
|
{
"evidence": [
"Rowan Atkinson",
"Blackadder"
],
"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:\nRowan Sebastian Atkinson is an Irish actor, comedian and writer who played the title role in a series of four period British sitcoms, plus several one-off installments, which originally aired on BBC One from 1983 to 1989.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Rowan Atkinson",
"Blackadder"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1700,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
75576
|
Julee Rosso is an American cook and food writer and she currently owns the Wickwood Inn in Saugatuck Michigan.
|
{
"evidence": [
"Julee Rosso_sentence_0",
"Julee Rosso_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:\nJulee Rosso is an American cook and food writer and she currently owns the Wickwood Inn in Saugatuck Michigan.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Julee Rosso_sentence_0",
"Julee Rosso_sentence_7"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1701,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.35",
"value": "?search="
},
{
"operation": "search",
"time": "28.587",
"value": "Julee Rosso"
},
{
"operation": "Now on",
"time": "29.284",
"value": "Julee Rosso"
},
{
"operation": "Highlighting",
"time": "49.466",
"value": "Julee Rosso_sentence_0"
},
{
"operation": "Highlighting",
"time": "67.801",
"value": "Julee Rosso_sentence_7"
},
{
"operation": "finish",
"time": "78.544",
"value": "finish"
}
]
}
|
90815
|
Giuseppe Canale, born in Rome in 1725, was instructed in engraving by Jacob Frey, and also frequented the school of the Cavalière Marco Benefial.
|
{
"evidence": [
"Giuseppe Canale_sentence_2",
"Giuseppe Canale_sentence_1",
"Giuseppe Canale_sentence_3"
],
"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:\nGiuseppe Canale, born in Rome in 1725, was instructed in engraving by Jacob Frey, and also frequented the school of the Cavalière Marco Benefial.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Giuseppe Canale_sentence_2",
"Giuseppe Canale_sentence_1",
"Giuseppe Canale_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1702,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.739",
"value": "?search="
},
{
"operation": "search",
"time": "10.425",
"value": "Giuseppe Canale"
},
{
"operation": "Now on",
"time": "12.481",
"value": "Giuseppe Canale"
},
{
"operation": "Highlighting",
"time": "17.156",
"value": "Giuseppe Canale_sentence_2"
},
{
"operation": "Highlighting",
"time": "19.245",
"value": "Giuseppe Canale_sentence_1"
},
{
"operation": "Highlighting",
"time": "22.809",
"value": "Giuseppe Canale_sentence_3"
},
{
"operation": "finish",
"time": "38.665",
"value": "finish"
}
]
}
|
801
|
Henry Bowyer Lane oversaw the construction of the incorporated city of Toronto's second city hall in 1844 and married his wife Lucy Anne Sharpe the following year.
|
{
"evidence": [
"Henry Bowyer Lane_sentence_6"
],
"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:\nHenry Bowyer Lane oversaw the construction of the incorporated city of Toronto's second city hall in 1844 and married his wife Lucy Anne Sharpe the following year.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Henry Bowyer Lane_sentence_6"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1703,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "search",
"time": "15.064",
"value": "Henry Bowyer Lane"
},
{
"operation": "Highlighting",
"time": "41.535",
"value": "Henry Bowyer Lane_sentence_4"
},
{
"operation": "Highlighting",
"time": "50.198",
"value": "Henry Bowyer Lane_sentence_6"
},
{
"operation": "Highlighting deleted",
"time": "53.813",
"value": "Henry Bowyer Lane_sentence_4"
},
{
"operation": "finish",
"time": "72.385",
"value": "finish"
}
]
}
|
71924
|
Armstrong Teasdale lost a Missouri Supreme Court ruling for Ameren in a wrongful death suit where their client Cave Consulting Group obtained a jury verdict of patent infringement.
|
{
"evidence": [
"Armstrong Teasdale_sentence_18"
],
"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:\nArmstrong Teasdale lost a Missouri Supreme Court ruling for Ameren in a wrongful death suit where their client Cave Consulting Group obtained a jury verdict of patent infringement.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Armstrong Teasdale_sentence_18"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1704,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.22",
"value": "?search="
},
{
"operation": "search",
"time": "8.627",
"value": "Armstrong Teasdale"
},
{
"operation": "Now on",
"time": "9.414",
"value": "Armstrong Teasdale"
},
{
"operation": "Page search",
"time": "19.279",
"value": "ameren"
},
{
"operation": "Highlighting",
"time": "27.678",
"value": "Armstrong Teasdale_sentence_18"
},
{
"operation": "finish",
"time": "44.302",
"value": "finish"
}
]
}
|
87976
|
During the Waddington By-Election 17 October 2002, Conservatives won more votes than Liberal Democrats which was founded in 1988.
|
{
"evidence": [
"Lincoln, England_sentence_2"
],
"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:\nDuring the Waddington By-Election 17 October 2002, Conservatives won more votes than Liberal Democrats which was founded in 1988.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Lincoln, England_sentence_2"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1705,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.387",
"value": "undefined"
},
{
"operation": "search",
"time": "27.017",
"value": "Waddington, Lancashire"
},
{
"operation": "Now on",
"time": "28.609",
"value": "Waddington, Lancashire"
},
{
"operation": "search",
"time": "99.418",
"value": "Waddington By-Election"
},
{
"operation": "Now on",
"time": "100.544",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "121.211",
"value": "Lincolnshire"
},
{
"operation": "Now on",
"time": "124.816",
"value": "Lincolnshire"
},
{
"operation": "Page search",
"time": "130.541",
"value": "waddin"
},
{
"operation": "Page search",
"time": "130.813",
"value": "waddin"
},
{
"operation": "Page search",
"time": "131.061",
"value": "waddin"
},
{
"operation": "Page search",
"time": "131.314",
"value": "waddin"
},
{
"operation": "Page search",
"time": "131.607",
"value": "waddin"
},
{
"operation": "Page search",
"time": "131.858",
"value": "waddin"
},
{
"operation": "Page search",
"time": "132.098",
"value": "waddin"
},
{
"operation": "back-button-clicked",
"time": "163.073",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "164.384",
"value": "undefined"
},
{
"operation": "search",
"time": "199.017",
"value": "2002 Waddington Election"
},
{
"operation": "Now on",
"time": "200.24",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "215.899",
"value": "Lincoln, England"
},
{
"operation": "Now on",
"time": "217.91",
"value": "Lincoln, England"
},
{
"operation": "Page search",
"time": "224.687",
"value": "waddin"
},
{
"operation": "Page search",
"time": "224.958",
"value": "waddin"
},
{
"operation": "Highlighting",
"time": "228.077",
"value": "Lincoln, England_sentence_2"
},
{
"operation": "hyperlink",
"time": "229.808",
"value": "Waddington"
},
{
"operation": "Now on",
"time": "230.872",
"value": "Waddington, Lincolnshire"
},
{
"operation": "finish",
"time": "249.763",
"value": "finish"
}
]
}
|
exfever_8610
|
The 2016 Summer Olympics is the first international multi-sport event participated by 207 nations.
|
{
"evidence": [
"2016 Summer Olympics",
"Refugee Olympic Team at the 2016 Summer Olympics"
],
"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 2016 Summer Olympics is the first international multi-sport event participated by 207 nations.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"2016 Summer Olympics",
"Refugee Olympic Team at the 2016 Summer Olympics"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1706,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
13837
|
Woodruff, a quarterback and punter at Kent State University, failed to join the University of Washington staff in Seattle as a graduate assistant under head coach Don James.
|
{
"evidence": [
"Jeff Woodruff_sentence_3"
],
"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:\nWoodruff, a quarterback and punter at Kent State University, failed to join the University of Washington staff in Seattle as a graduate assistant under head coach Don James.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Jeff Woodruff_sentence_3"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1707,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0",
"value": "?search="
},
{
"operation": "search",
"time": "0.014",
"value": "Don James"
},
{
"operation": "Now on",
"time": "0.015",
"value": "Don James"
},
{
"operation": "hyperlink",
"time": "0.018",
"value": "Don James (American football)"
},
{
"operation": "Now on",
"time": "0.019",
"value": "Don James (American football)"
},
{
"operation": "search",
"time": "0.038",
"value": "kent state university football"
},
{
"operation": "Now on",
"time": "0.04",
"value": "undefined"
},
{
"operation": "search",
"time": "0.053",
"value": "kent state university football woodruff"
},
{
"operation": "Now on",
"time": "0.054",
"value": "undefined"
},
{
"operation": "search",
"time": "0.088",
"value": "Jeff Woodruff"
},
{
"operation": "Now on",
"time": "0.089",
"value": "Jeff Woodruff"
},
{
"operation": "Highlighting",
"time": "0.1",
"value": "Jeff Woodruff_sentence_2"
},
{
"operation": "Highlighting",
"time": "0.113",
"value": "Jeff Woodruff_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "0.119",
"value": "Jeff Woodruff_sentence_2"
},
{
"operation": "finish",
"time": "0.13",
"value": "finish"
},
{
"operation": "Now on",
"time": "1619468431.735",
"value": "Jeff Woodruff"
},
{
"operation": "Highlighting",
"time": "1619468441.881",
"value": "Jeff Woodruff_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "1619468444.812",
"value": "Jeff Woodruff_sentence_2"
},
{
"operation": "finish",
"time": "1619468448.147",
"value": "finish"
}
]
}
|
exfever_532
|
Thomas Jacob Black is an American actor who voices the giant panda named Po from Kung Fu Panda films by an American animation studio that produces animated films and television programs and is a subsidiary of Universal Pictures, a division of NBCUniversal.
|
{
"evidence": [
"Jack Black",
"DreamWorks Animation"
],
"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:\nThomas Jacob Black is an American actor who voices the giant panda named Po from Kung Fu Panda films by an American animation studio that produces animated films and television programs and is a subsidiary of Universal Pictures, a division of NBCUniversal.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Jack Black",
"DreamWorks Animation"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1708,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6206
|
Radiohead is the first band to have worked with producer Nigel Godrich.
|
{
"evidence": [
"Radiohead",
"Rock music"
],
"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:\nRadiohead is the first band to have worked with producer Nigel Godrich.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Radiohead",
"Rock music"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1709,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
5303
|
Indalecio Bizcarrondo, also known as Bilintx, was a Basque improvisational poet or bertsolari living in Spain who associated in the liberal circles of the city, with the likes of Serafin Baroja and Manterola, in the period extending between the two main Carlist Wars.
|
{
"evidence": [
"Bilintx_sentence_0",
"Bilintx_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:\nIndalecio Bizcarrondo, also known as Bilintx, was a Basque improvisational poet or bertsolari living in Spain who associated in the liberal circles of the city, with the likes of Serafin Baroja and Manterola, in the period extending between the two main Carlist Wars.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Bilintx_sentence_0",
"Bilintx_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1710,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.005",
"value": "?search="
},
{
"operation": "search",
"time": "32.661",
"value": "Indalecio Bizcarrondo"
},
{
"operation": "Now on",
"time": "33.449",
"value": "Bilintx"
},
{
"operation": "Highlighting",
"time": "38.764",
"value": "Bilintx_sentence_0"
},
{
"operation": "Highlighting",
"time": "63.127",
"value": "Bilintx_sentence_1"
},
{
"operation": "finish",
"time": "169.111",
"value": "finish"
}
]
}
|
63947
|
In 1599, Matsura Shigenobu erected a castle in Hirado Domain, and it still stands today in it's original form.
|
{
"evidence": [
"Hirado Castle_sentence_7",
"Hirado Castle_sentence_8",
"Hirado Castle_sentence_13"
],
"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:\nIn 1599, Matsura Shigenobu erected a castle in Hirado Domain, and it still stands today in it's original form.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Hirado Castle_sentence_7",
"Hirado Castle_sentence_8",
"Hirado Castle_sentence_13"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1711,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.332",
"value": "?search="
},
{
"operation": "search",
"time": "59.387",
"value": "Contains...Matsura Shigenobu"
},
{
"operation": "Now on",
"time": "88.258",
"value": "?search="
},
{
"operation": "search",
"time": "106.314",
"value": "hirado castle"
},
{
"operation": "Now on",
"time": "108.718",
"value": "Hirado Castle"
},
{
"operation": "Highlighting",
"time": "118.433",
"value": "Hirado Castle_sentence_7"
},
{
"operation": "Highlighting",
"time": "137.614",
"value": "Hirado Castle_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "151.354",
"value": "Hirado Castle_sentence_2"
},
{
"operation": "Highlighting",
"time": "155.41",
"value": "Hirado Castle_sentence_8"
},
{
"operation": "Highlighting",
"time": "157.495",
"value": "Hirado Castle_sentence_13"
},
{
"operation": "finish",
"time": "162.839",
"value": "finish"
}
]
}
|
exfever_142
|
"See You Again" by Miley Cyrus was produced by Tim James and a Canadian songwriter, record producer, music producer, and businesswoman.
|
{
"evidence": [
"See You Again (Miley Cyrus song)",
"Antonina Armato"
],
"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\"See You Again\" by Miley Cyrus was produced by Tim James and a Canadian songwriter, record producer, music producer, and businesswoman.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"See You Again (Miley Cyrus song)",
"Antonina Armato"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1712,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3073
|
Music is the art of arranging sounds in time through the elements such as a perceptual property of sounds that is a major auditory attribute of musical tones, along with an amount of time or how long or short a note, phrase, section, or composition lasts.
|
{
"evidence": [
"Music",
"Pitch (music)",
"Duration (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:\nMusic is the art of arranging sounds in time through the elements such as a perceptual property of sounds that is a major auditory attribute of musical tones, along with an amount of time or how long or short a note, phrase, section, or composition lasts.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Music",
"Pitch (music)",
"Duration (music)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1713,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1441
|
Bonnie Hunt is an American actress whose film roles include a 1988 American road comedy-drama film directed by Barry Levinson.
|
{
"evidence": [
"Bonnie Hunt",
"Rain Man"
],
"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:\nBonnie Hunt is an American actress whose film roles include a 1988 American road comedy-drama film directed by Barry Levinson.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Bonnie Hunt",
"Rain Man"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1714,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
46413
|
The Royal Geelong Yacht Club, which received its royal patronage in 1924, is built in the 1980s on Gold Coast, Australia.
|
{
"evidence": [
"Royal Geelong Yacht Club_sentence_3"
],
"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 Royal Geelong Yacht Club, which received its royal patronage in 1924, is built in the 1980s on Gold Coast, Australia.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Royal Geelong Yacht Club_sentence_3"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1715,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.713",
"value": "?search="
},
{
"operation": "search",
"time": "13.147",
"value": "Royal Geelong Yacht Club"
},
{
"operation": "Now on",
"time": "13.822",
"value": "Royal Geelong Yacht Club"
},
{
"operation": "Highlighting",
"time": "20.488",
"value": "Royal Geelong Yacht Club_sentence_1"
},
{
"operation": "Highlighting",
"time": "55.643",
"value": "Royal Geelong Yacht Club_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "69.048",
"value": "Royal Geelong Yacht Club_sentence_1"
},
{
"operation": "finish",
"time": "77.694",
"value": "finish"
}
]
}
|
exfever_8473
|
John Wayne Gacy is the first serial killer to use the method of disguising into a clown to lure young men before killing them.
|
{
"evidence": [
"John Wayne Gacy",
"Handcuffs"
],
"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:\nJohn Wayne Gacy is the first serial killer to use the method of disguising into a clown to lure young men before killing them.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"John Wayne Gacy",
"Handcuffs"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1716,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7783
|
John Woo is the most famous film director in Hongkong.
|
{
"evidence": [
"John Woo",
"Silver Bauhinia Star"
],
"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:\nJohn Woo is the most famous film director in Hongkong.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"John Woo",
"Silver Bauhinia Star"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1717,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
85405
|
There are 30 numbered country roads in Haldimand County
|
{
"evidence": [
"Haldimand_sentence_2"
],
"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:\nThere are 30 numbered country roads in Haldimand County\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Haldimand_sentence_2"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1718,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.84",
"value": "?search="
},
{
"operation": "search",
"time": "25.283",
"value": "Contains...country roads in Haldimand County"
},
{
"operation": "Now on",
"time": "27.256",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "31.486",
"value": "Haldimand"
},
{
"operation": "Now on",
"time": "32.416",
"value": "Haldimand"
},
{
"operation": "Highlighting",
"time": "80.851",
"value": "Haldimand_sentence_2"
},
{
"operation": "finish",
"time": "106.44",
"value": "finish"
}
]
}
|
exfever_758
|
Trevor Noah is a South African television host who succeeded an American comedian who was a long-time host of The Daily Show.
|
{
"evidence": [
"Trevor Noah",
"Jon Stewart"
],
"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:\nTrevor Noah is a South African television host who succeeded an American comedian who was a long-time host of The Daily Show.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Trevor Noah",
"Jon Stewart"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1719,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
41887
|
The first game of the 1998 Philadelphia Phillies season was a win against the New York Mets.
|
{
"evidence": [
"1998 Philadelphia Phillies season_sentence_1",
"1998 Philadelphia Phillies season_sentence_3"
],
"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 first game of the 1998 Philadelphia Phillies season was a win against the New York Mets.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"1998 Philadelphia Phillies season_sentence_1",
"1998 Philadelphia Phillies season_sentence_3"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1720,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.197",
"value": "?search="
},
{
"operation": "search",
"time": "9.643",
"value": "1998 Philadelphia Phillies season"
},
{
"operation": "Now on",
"time": "11.047",
"value": "1998 Philadelphia Phillies season"
},
{
"operation": "Highlighting",
"time": "19.429",
"value": "1998 Philadelphia Phillies season_sentence_1"
},
{
"operation": "Highlighting",
"time": "37.339",
"value": "1998 Philadelphia Phillies season_sentence_3"
},
{
"operation": "finish",
"time": "74.589",
"value": "finish"
}
]
}
|
exfever_5988
|
Tennis is a racket sport that can be played against a single or doubles opponent on a firm rectangular surface with a low net stretched across the center.
|
{
"evidence": [
"Tennis",
"Tennis court"
],
"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:\nTennis is a racket sport that can be played against a single or doubles opponent on a firm rectangular surface with a low net stretched across the center.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Tennis",
"Tennis court"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1721,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
709
|
In the list of Unification of Norway (860–872), Vestfold (current electoral district in Eastern Norway) gained four victories while Viking gained three victories.
|
{
"evidence": [
"Vestfold_sentence_0",
"Vestfold_sentence_25"
],
"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 the list of Unification of Norway (860–872), Vestfold (current electoral district in Eastern Norway) gained four victories while Viking gained three victories.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Vestfold_sentence_0",
"Vestfold_sentence_25"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1722,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "search",
"time": "10.856",
"value": "Vestfold"
},
{
"operation": "Highlighting",
"time": "18.277",
"value": "Vestfold_sentence_0"
},
{
"operation": "hyperlink",
"time": "23.683",
"value": "2.1 Viking history Vestfold_section_2"
},
{
"operation": "Highlighting",
"time": "25.897",
"value": "Vestfold_sentence_25"
},
{
"operation": "search",
"time": "42.013",
"value": "unification of norway"
},
{
"operation": "finish",
"time": "55.991",
"value": "finish"
}
]
}
|
exfever_280
|
In Dubious Battle features an ensemble cast, including an American actor and a recipient of an annual award to honour the best British and international contributions to film.
|
{
"evidence": [
"In Dubious Battle (film)",
"Robert Duvall",
"British Academy Film Awards"
],
"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:\nIn Dubious Battle features an ensemble cast, including an American actor and a recipient of an annual award to honour the best British and international contributions to film.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"In Dubious Battle (film)",
"Robert Duvall",
"British Academy Film Awards"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1723,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
1820
|
The 1994 Seattle Seahawks season was the team's 30th season with the NFL and was coached by Tom Flores.
|
{
"evidence": [
"1994 Seattle Seahawks season_sentence_0",
"1994 Seattle Seahawks season_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 1994 Seattle Seahawks season was the team's 30th season with the NFL and was coached by Tom Flores.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"1994 Seattle Seahawks season_sentence_0",
"1994 Seattle Seahawks season_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1724,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "search",
"time": "25.965",
"value": "Contains...The 1994 Seattle Seahawks"
},
{
"operation": "hyperlink",
"time": "42.762",
"value": "1994 Seattle Seahawks season"
},
{
"operation": "Highlighting",
"time": "49.371",
"value": "1994 Seattle Seahawks season_sentence_0"
},
{
"operation": "Highlighting",
"time": "56.655",
"value": "1994 Seattle Seahawks season_sentence_1"
},
{
"operation": "finish",
"time": "81.547",
"value": "finish"
}
]
}
|
exfever_2003
|
Life Is Peachy ranked 3 on US Billboard 200.
|
{
"evidence": [
"Life Is Peachy",
"Korn"
],
"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:\nLife Is Peachy ranked 3 on US Billboard 200.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Life Is Peachy",
"Korn"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1725,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8263
|
The Daily Show was the longest satirical news television program in America.
|
{
"evidence": [
"The Daily Show",
"Trevor Noah"
],
"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 Daily Show was the longest satirical news television program in America.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Daily Show",
"Trevor Noah"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1726,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8387
|
Schwarzenegger was the first actor named one of the most influential people in the world by Time magazine.
|
{
"evidence": [
"Arnold Schwarzenegger",
"Weight training"
],
"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:\nSchwarzenegger was the first actor named one of the most influential people in the world by Time magazine.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Arnold Schwarzenegger",
"Weight training"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1727,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8754
|
Lana Wachowski and Lilly Wachowski achieved fame with their second film The Matrix, a major box office success for which they won the annual awards given by the American Academy of Science Fiction, Fantasy & Horror Films.
|
{
"evidence": [
"The Wachowskis",
"Saturn Award for Best Director"
],
"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:\nLana Wachowski and Lilly Wachowski achieved fame with their second film The Matrix, a major box office success for which they won the annual awards given by the American Academy of Science Fiction, Fantasy & Horror Films.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Wachowskis",
"Saturn Award for Best Director"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1728,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
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": 1729,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
30361
|
Teldenia alba Adults are greyish white, both wings with a transverse discal zigzag band, composed of purplish lunules with black-speckled inner borders,Beyond the lunular band is a very indistinct pale purplish-brown lunular line and a marginal row of black dots.
|
{
"evidence": [
"Teldenia alba_sentence_4",
"Teldenia alba_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:\nTeldenia alba Adults are greyish white, both wings with a transverse discal zigzag band, composed of purplish lunules with black-speckled inner borders,Beyond the lunular band is a very indistinct pale purplish-brown lunular line and a marginal row of black dots.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Teldenia alba_sentence_4",
"Teldenia alba_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1730,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.063",
"value": "?search="
},
{
"operation": "search",
"time": "6.085",
"value": "Teldenia alba"
},
{
"operation": "Now on",
"time": "7.145",
"value": "Teldenia alba"
},
{
"operation": "Highlighting",
"time": "10.44",
"value": "Teldenia alba_sentence_4"
},
{
"operation": "Highlighting",
"time": "20.153",
"value": "Teldenia alba_sentence_5"
},
{
"operation": "finish",
"time": "34.364",
"value": "finish"
}
]
}
|
exfever_845
|
Jane Austen's Emma is an adaptation of the 1815 novel of the same name, starring an English actress and model who has appeared in British costume dramas, including a 2000 period romantic drama film directed by James Ivory.
|
{
"evidence": [
"Emma (1996 TV film)",
"Kate Beckinsale",
"The Golden Bowl (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:\nJane Austen's Emma is an adaptation of the 1815 novel of the same name, starring an English actress and model who has appeared in British costume dramas, including a 2000 period romantic drama film directed by James Ivory.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Emma (1996 TV film)",
"Kate Beckinsale",
"The Golden Bowl (film)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1731,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
54498
|
The 1996 National Football League season was the Seattle Seahawks' 15th season to participate; then the team was rented to Paul Walker in 2005.
|
{
"evidence": [
"1996 NFL season_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 1996 National Football League season was the Seattle Seahawks' 15th season to participate; then the team was rented to Paul Walker in 2005.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"1996 NFL season_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1732,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.381",
"value": "?search="
},
{
"operation": "search",
"time": "6.979",
"value": "1996 National Football League"
},
{
"operation": "Now on",
"time": "9.615",
"value": "undefined"
},
{
"operation": "search",
"time": "73.832",
"value": "1996 Seattle Seahawks"
},
{
"operation": "Now on",
"time": "76.32",
"value": "1996 Seattle Seahawks season"
},
{
"operation": "search",
"time": "105.991",
"value": "Seattle Seahawks"
},
{
"operation": "Now on",
"time": "114.193",
"value": "1996 Seattle Seahawks season"
},
{
"operation": "search",
"time": "135.566",
"value": "Seattle Seahawks"
},
{
"operation": "Now on",
"time": "137.548",
"value": "Seattle Seahawks"
},
{
"operation": "Page search",
"time": "141.768",
"value": "1996"
},
{
"operation": "Page search",
"time": "190.788",
"value": "paul walker"
},
{
"operation": "Page search",
"time": "207.556",
"value": "rented"
},
{
"operation": "Page search",
"time": "219.721",
"value": "walker"
},
{
"operation": "search",
"time": "230.865",
"value": "team was rented to Paul Walker in 2005"
},
{
"operation": "Now on",
"time": "233.56",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "244.703",
"value": "H. G. Wells"
},
{
"operation": "Now on",
"time": "246.347",
"value": "H. G. Wells"
},
{
"operation": "search",
"time": "295.631",
"value": "Contains... Seattle Seahawks' 15th season to participate"
},
{
"operation": "Now on",
"time": "297.811",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "304.325",
"value": "Week 5"
},
{
"operation": "Now on",
"time": "314.801",
"value": "2011 New York Giants season#Week 5"
},
{
"operation": "search",
"time": "334.573",
"value": "Seattle seahawks seasons"
},
{
"operation": "Now on",
"time": "337.496",
"value": "List of Seattle Seahawks seasons"
},
{
"operation": "Page search",
"time": "343.616",
"value": "15th"
},
{
"operation": "hyperlink",
"time": "350.006",
"value": "History of the Seattle Seahawks"
},
{
"operation": "Now on",
"time": "352.521",
"value": "History of the Seattle Seahawks"
},
{
"operation": "Page search",
"time": "359.538",
"value": "15th season"
},
{
"operation": "Page search",
"time": "375.421",
"value": "walker"
},
{
"operation": "Highlighting",
"time": "386.299",
"value": "History of the Seattle Seahawks_sentence_182"
},
{
"operation": "Highlighting",
"time": "386.301",
"value": "History of the Seattle Seahawks_title"
},
{
"operation": "Highlighting",
"time": "386.304",
"value": "History of the Seattle Seahawks_section_9"
},
{
"operation": "Highlighting",
"time": "386.304",
"value": "History of the Seattle Seahawks_section_14"
},
{
"operation": "Highlighting deleted",
"time": "387.116",
"value": "History of the Seattle Seahawks_sentence_182"
},
{
"operation": "Highlighting deleted",
"time": "387.119",
"value": "History of the Seattle Seahawks_title"
},
{
"operation": "Highlighting deleted",
"time": "387.122",
"value": "History of the Seattle Seahawks_section_9"
},
{
"operation": "Highlighting deleted",
"time": "387.124",
"value": "History of the Seattle Seahawks_section_14"
},
{
"operation": "search",
"time": "396.622",
"value": "1996 National Football League season"
},
{
"operation": "Now on",
"time": "400.328",
"value": "1996 NFL season"
},
{
"operation": "Highlighting",
"time": "425.203",
"value": "1996 NFL season_sentence_0"
},
{
"operation": "finish",
"time": "430.178",
"value": "finish"
}
]
}
|
exfever_3143
|
Paul McCartney has won a Grammy Award for his song "Pipes of Peace".
|
{
"evidence": [
"Paul McCartney",
"Electronica"
],
"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:\nPaul McCartney has won a Grammy Award for his song \"Pipes of Peace\".\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Paul McCartney",
"Electronica"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1733,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
89662
|
Kafai was born in South Korea and studied in Germany, France and she worked in the United States at the MIT Media Laboratory.
|
{
"evidence": [
"Yasmin Kafai_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:\nKafai was born in South Korea and studied in Germany, France and she worked in the United States at the MIT Media Laboratory.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Yasmin Kafai_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1734,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.436",
"value": "?search="
},
{
"operation": "search",
"time": "14.43",
"value": "Contains...Kafai"
},
{
"operation": "Now on",
"time": "15.801",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "28.209",
"value": "Yasmin (given name)"
},
{
"operation": "Now on",
"time": "29.306",
"value": "Yasmin (given name)"
},
{
"operation": "search",
"time": "36.248",
"value": "Kafaiyeh"
},
{
"operation": "Now on",
"time": "37.394",
"value": "Tajabad, Rafsanjan"
},
{
"operation": "search",
"time": "47.561",
"value": "Contains...Kafai"
},
{
"operation": "Now on",
"time": "48.268",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "54.341",
"value": "Mitchel Resnick"
},
{
"operation": "Now on",
"time": "55.322",
"value": "Mitchel Resnick"
},
{
"operation": "search",
"time": "61.307",
"value": "Yasmin Kafai"
},
{
"operation": "Now on",
"time": "62.177",
"value": "Yasmin Kafai"
},
{
"operation": "Highlighting",
"time": "66.654",
"value": "Yasmin Kafai_sentence_1"
},
{
"operation": "finish",
"time": "87.613",
"value": "finish"
}
]
}
|
49465
|
The Enlist Weed Control System, developed by Dow AgroSciences, part of Dow Chemical Company, is an agricultural system that includes seeds for organic crops that are resistant to Enlist and the Enlist herbicide.
|
{
"evidence": [
"Enlist Weed Control System_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:\nThe Enlist Weed Control System, developed by Dow AgroSciences, part of Dow Chemical Company, is an agricultural system that includes seeds for organic crops that are resistant to Enlist and the Enlist herbicide.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Enlist Weed Control System_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1735,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.514",
"value": "?search="
},
{
"operation": "search",
"time": "12.962",
"value": "Enlist Weed Control System"
},
{
"operation": "Now on",
"time": "14.362",
"value": "Enlist Weed Control System"
},
{
"operation": "Highlighting",
"time": "31.87",
"value": "Enlist Weed Control System_sentence_1"
},
{
"operation": "Highlighting",
"time": "40.988",
"value": "Enlist Weed Control System_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "44.325",
"value": "Enlist Weed Control System_sentence_1"
},
{
"operation": "Highlighting",
"time": "44.794",
"value": "Enlist Weed Control System_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "73.219",
"value": "Enlist Weed Control System_sentence_1"
},
{
"operation": "finish",
"time": "95.674",
"value": "finish"
}
]
}
|
44309
|
Silhadi rose to positions in his military career during Sultan Mahmood Khilji II's power and became an important person in the northeastern Malwa region.
|
{
"evidence": [
"Silhadi_sentence_10",
"Silhadi_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:\nSilhadi rose to positions in his military career during Sultan Mahmood Khilji II's power and became an important person in the northeastern Malwa region.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Silhadi_sentence_10",
"Silhadi_sentence_9"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1736,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.596",
"value": "?search="
},
{
"operation": "search",
"time": "22.149",
"value": "Silhadi"
},
{
"operation": "Now on",
"time": "22.886",
"value": "Silhadi"
},
{
"operation": "Page search",
"time": "48.452",
"value": "khilji"
},
{
"operation": "Highlighting",
"time": "52.54",
"value": "Silhadi_sentence_9"
},
{
"operation": "Highlighting",
"time": "70.218",
"value": "Silhadi_sentence_10"
},
{
"operation": "Page search",
"time": "88.194",
"value": "malta"
},
{
"operation": "page-search-reset",
"time": "97.577",
"value": "page-search-reset"
},
{
"operation": "Highlighting deleted",
"time": "112.514",
"value": "Silhadi_sentence_9"
},
{
"operation": "Highlighting",
"time": "113.354",
"value": "Silhadi_sentence_9"
},
{
"operation": "finish",
"time": "157.528",
"value": "finish"
}
]
}
|
58753
|
Pentraxin-related protein PTX3, also known as TNF-inducible gene 14 protein, is a member of the pentraxin superfamily or phylogenetically conserved humoral mediators of innate immunity, a protein that in humans is encoded by the PTX3 gene.
|
{
"evidence": [
"PTX3_sentence_2",
"PTX3_sentence_3",
"PTX3_sentence_17"
],
"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:\nPentraxin-related protein PTX3, also known as TNF-inducible gene 14 protein, is a member of the pentraxin superfamily or phylogenetically conserved humoral mediators of innate immunity, a protein that in humans is encoded by the PTX3 gene.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"PTX3_sentence_2",
"PTX3_sentence_3",
"PTX3_sentence_17"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1737,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.612",
"value": "?search="
},
{
"operation": "search",
"time": "17.099",
"value": "PTX3"
},
{
"operation": "Now on",
"time": "18.022",
"value": "PTX3"
},
{
"operation": "Highlighting",
"time": "45.327",
"value": "PTX3_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "45.703",
"value": "PTX3_sentence_1"
},
{
"operation": "Highlighting",
"time": "46.167",
"value": "PTX3_sentence_2"
},
{
"operation": "Highlighting",
"time": "53.377",
"value": "PTX3_sentence_3"
},
{
"operation": "hyperlink",
"time": "64.314",
"value": "pentraxin"
},
{
"operation": "Now on",
"time": "65.043",
"value": "Pentraxins"
},
{
"operation": "back-button-clicked",
"time": "71.491",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "71.574",
"value": "PTX3"
},
{
"operation": "Highlighting",
"time": "99.616",
"value": "PTX3_sentence_17"
},
{
"operation": "search",
"time": "154.634",
"value": "Phylogenetically conserved"
},
{
"operation": "Now on",
"time": "155.913",
"value": "Conserved sequence"
},
{
"operation": "finish",
"time": "177.513",
"value": "finish"
}
]
}
|
exfever_4879
|
Christopher James Evans is an English television presenter who his long-running Drivetime program in April 2006, before moving in 2010 to host a radio program hosted by broadcaster Chris Evans in the United Kingdom, meanwhile started the second incarnation of the show on a national Hot Adult Contemporary radio station in the United Kingdom that launched on 30 March 2016, owned by Wireless Group.
|
{
"evidence": [
"Chris Evans (presenter)",
"The Chris Evans Breakfast Show",
"Virgin Radio UK"
],
"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:\nChristopher James Evans is an English television presenter who his long-running Drivetime program in April 2006, before moving in 2010 to host a radio program hosted by broadcaster Chris Evans in the United Kingdom, meanwhile started the second incarnation of the show on a national Hot Adult Contemporary radio station in the United Kingdom that launched on 30 March 2016, owned by Wireless Group.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Chris Evans (presenter)",
"The Chris Evans Breakfast Show",
"Virgin Radio UK"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1738,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
44716
|
Kollam Canal is a part of Kollam Canal development project that is expected to meet the water transportation and irrigation needs of the people of Kollam Urban Agglomeration, the 14th largest urban agglomeration in South India and 7th largest in Kerala, in future.
|
{
"evidence": [
"Kollam Canal_sentence_8",
"Kollam Canal_sentence_11",
"Kollam metropolitan area_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:\nKollam Canal is a part of Kollam Canal development project that is expected to meet the water transportation and irrigation needs of the people of Kollam Urban Agglomeration, the 14th largest urban agglomeration in South India and 7th largest in Kerala, in future.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Kollam Canal_sentence_8",
"Kollam Canal_sentence_11",
"Kollam metropolitan area_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1739,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.615",
"value": "?search="
},
{
"operation": "search",
"time": "18.041",
"value": "Kollam Canal"
},
{
"operation": "Now on",
"time": "19.156",
"value": "Kollam Canal"
},
{
"operation": "Highlighting",
"time": "187.981",
"value": "Kollam Canal_sentence_2"
},
{
"operation": "Highlighting",
"time": "187.983",
"value": "Kollam Canal_section_0"
},
{
"operation": "Highlighting",
"time": "221.625",
"value": "Kollam Canal_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "223.352",
"value": "Kollam Canal_sentence_1"
},
{
"operation": "Highlighting",
"time": "225.808",
"value": "Kollam Canal_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "231.309",
"value": "Kollam Canal_sentence_1"
},
{
"operation": "Highlighting",
"time": "242.359",
"value": "Kollam Canal_sentence_8"
},
{
"operation": "Highlighting",
"time": "248.027",
"value": "Kollam Canal_sentence_8"
},
{
"operation": "Highlighting",
"time": "257.555",
"value": "Kollam Canal_sentence_11"
},
{
"operation": "hyperlink",
"time": "283.997",
"value": "Kollam Urban Agglomeration"
},
{
"operation": "Now on",
"time": "285.301",
"value": "Kollam metropolitan area"
},
{
"operation": "Highlighting",
"time": "288.653",
"value": "Kollam metropolitan area_sentence_2"
},
{
"operation": "finish",
"time": "331.067",
"value": "finish"
}
]
}
|
92320
|
Siuslaw National Forest contains Marys Peak and has very few aquatic habitats.
|
{
"evidence": [
"Siuslaw National Forest_sentence_7"
],
"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:\nSiuslaw National Forest contains Marys Peak and has very few aquatic habitats.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Siuslaw National Forest_sentence_7"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1740,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.231",
"value": "?search="
},
{
"operation": "search",
"time": "15.972",
"value": "Siuslaw National Forest"
},
{
"operation": "Now on",
"time": "16.516",
"value": "Siuslaw National Forest"
},
{
"operation": "Highlighting",
"time": "151.819",
"value": "Siuslaw National Forest_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "154.57",
"value": "Siuslaw National Forest_sentence_2"
},
{
"operation": "Highlighting",
"time": "156.203",
"value": "Siuslaw National Forest_sentence_6"
},
{
"operation": "Highlighting",
"time": "160.179",
"value": "Siuslaw National Forest_sentence_7"
},
{
"operation": "Highlighting deleted",
"time": "161.043",
"value": "Siuslaw National Forest_sentence_6"
},
{
"operation": "finish",
"time": "167.179",
"value": "finish"
}
]
}
|
66464
|
There are four settlements in the Federation of Bosnia and Herzegovina which include Njeganovići in Bileća, Njemanicain Stari Grad, Sarajevo, Njivak in Gradačac and Njuhe in Foča.
|
{
"evidence": [
"Federation of Bosnia and Herzegovina_sentence_1",
"Federation of Bosnia and Herzegovina_sentence_2"
],
"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:\nThere are four settlements in the Federation of Bosnia and Herzegovina which include Njeganovići in Bileća, Njemanicain Stari Grad, Sarajevo, Njivak\tin Gradačac and Njuhe in Foča.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Federation of Bosnia and Herzegovina_sentence_1",
"Federation of Bosnia and Herzegovina_sentence_2"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1741,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.23",
"value": "?search="
},
{
"operation": "search",
"time": "32.339",
"value": "settlements in the Federation of Bosnia and Herzegovina"
},
{
"operation": "Now on",
"time": "35.485",
"value": "undefined"
},
{
"operation": "search",
"time": "55.619",
"value": "Bosnia and Herzegovina"
},
{
"operation": "Now on",
"time": "74.619",
"value": "Bosnia and Herzegovina"
},
{
"operation": "hyperlink",
"time": "132.794",
"value": "Bosnia (region)"
},
{
"operation": "Now on",
"time": "133.344",
"value": "Bosnia (region)"
},
{
"operation": "hyperlink",
"time": "137.386",
"value": "4 Subregions Bosnia (region)_section_3"
},
{
"operation": "back-button-clicked",
"time": "150.531",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "150.721",
"value": "Bosnia and Herzegovina"
},
{
"operation": "Page search",
"time": "159.348",
"value": "bileca"
},
{
"operation": "Page search",
"time": "162.693",
"value": "bile*"
},
{
"operation": "Page search",
"time": "178.359",
"value": "Sarajevo"
},
{
"operation": "Highlighting",
"time": "234.595",
"value": "Bosnia and Herzegovina_sentence_287"
},
{
"operation": "Highlighting",
"time": "234.596",
"value": "Bosnia and Herzegovina_title"
},
{
"operation": "Highlighting",
"time": "234.596",
"value": "Bosnia and Herzegovina_section_14"
},
{
"operation": "Highlighting",
"time": "234.596",
"value": "Bosnia and Herzegovina_section_18"
},
{
"operation": "Highlighting deleted",
"time": "246.018",
"value": "Bosnia and Herzegovina_sentence_287"
},
{
"operation": "Highlighting deleted",
"time": "246.018",
"value": "Bosnia and Herzegovina_title"
},
{
"operation": "Highlighting deleted",
"time": "246.019",
"value": "Bosnia and Herzegovina_section_14"
},
{
"operation": "Highlighting deleted",
"time": "246.02",
"value": "Bosnia and Herzegovina_section_18"
},
{
"operation": "Highlighting",
"time": "247.37",
"value": "Bosnia and Herzegovina_sentence_288"
},
{
"operation": "Highlighting",
"time": "247.37",
"value": "Bosnia and Herzegovina_title"
},
{
"operation": "Highlighting",
"time": "247.37",
"value": "Bosnia and Herzegovina_section_14"
},
{
"operation": "Highlighting",
"time": "247.371",
"value": "Bosnia and Herzegovina_section_18"
},
{
"operation": "Highlighting deleted",
"time": "247.802",
"value": "Bosnia and Herzegovina_sentence_288"
},
{
"operation": "Highlighting deleted",
"time": "247.802",
"value": "Bosnia and Herzegovina_title"
},
{
"operation": "Highlighting deleted",
"time": "247.803",
"value": "Bosnia and Herzegovina_section_14"
},
{
"operation": "Highlighting deleted",
"time": "247.803",
"value": "Bosnia and Herzegovina_section_18"
},
{
"operation": "hyperlink",
"time": "275.881",
"value": "Outline of Bosnia and Herzegovina"
},
{
"operation": "Now on",
"time": "276.785",
"value": "Outline of Bosnia and Herzegovina"
},
{
"operation": "Highlighting",
"time": "294.025",
"value": "Outline of Bosnia and Herzegovina_sentence_2"
},
{
"operation": "Highlighting",
"time": "294.025",
"value": "Outline of Bosnia and Herzegovina_title"
},
{
"operation": "hyperlink",
"time": "301.393",
"value": "2.2 Regions of Bosnia and Herzegovina Outline of Bosnia and Herzegovina_section_4"
},
{
"operation": "hyperlink",
"time": "309.281",
"value": "Regions of Bosnia and Herzegovina"
},
{
"operation": "Now on",
"time": "309.816",
"value": "Regions of Bosnia and Herzegovina"
},
{
"operation": "back-button-clicked",
"time": "320.89",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "320.976",
"value": "Outline of Bosnia and Herzegovina#Regions of Bosnia and Herzegovina Outline of Bosnia and Herzegovina section 4"
},
{
"operation": "back-button-clicked",
"time": "322.545",
"value": "back-button-clicked"
},
{
"operation": "Page search",
"time": "326.889",
"value": "settlements"
},
{
"operation": "search",
"time": "368.217",
"value": "Njeganvovici in bileca"
},
{
"operation": "Now on",
"time": "369.507",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "373.041",
"value": "njeganvovici in bilecik"
},
{
"operation": "Now on",
"time": "375.028",
"value": "undefined"
},
{
"operation": "search",
"time": "385.801",
"value": "four settlements"
},
{
"operation": "Now on",
"time": "386.937",
"value": "undefined"
},
{
"operation": "search",
"time": "395.937",
"value": "Nijuhe"
},
{
"operation": "Now on",
"time": "399.962",
"value": "undefined"
},
{
"operation": "search",
"time": "418.994",
"value": "Federation of Bosnia and Herzegovina"
},
{
"operation": "Now on",
"time": "419.905",
"value": "Federation of Bosnia and Herzegovina"
},
{
"operation": "Highlighting",
"time": "433.721",
"value": "Federation of Bosnia and Herzegovina_sentence_1"
},
{
"operation": "Highlighting",
"time": "434.993",
"value": "Federation of Bosnia and Herzegovina_sentence_2"
},
{
"operation": "Page search",
"time": "446.249",
"value": "settlement"
},
{
"operation": "hyperlink",
"time": "475.809",
"value": "Inter-Entity Boundary Line"
},
{
"operation": "Now on",
"time": "479.164",
"value": "Inter-Entity Boundary Line"
},
{
"operation": "finish",
"time": "503.281",
"value": "finish"
}
]
}
|
exfever_3018
|
Catching Fire is the second book in a 2008 dystopian novel by an American television writer and author.
|
{
"evidence": [
"Catching Fire",
"The Hunger Games (novel)",
"Suzanne Collins"
],
"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:\nCatching Fire is the second book in a 2008 dystopian novel by an American television writer and author.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Catching Fire",
"The Hunger Games (novel)",
"Suzanne Collins"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1742,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
74198
|
Larry Ross was third in the British League World Riders Championship in 1980, and rode Long Track the next year; from 1977 to 1986, he rode in eight finals of the World Pairs and was runnerup with Ivan Mauger in 1978 and 1981.
|
{
"evidence": [
"Larry Ross_sentence_6"
],
"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:\nLarry Ross was third in the British League World Riders Championship in 1980, and rode Long Track the next year; from 1977 to 1986, he rode in eight finals of the World Pairs and was runnerup with Ivan Mauger in 1978 and 1981.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Larry Ross_sentence_6"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1743,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.261",
"value": "?search="
},
{
"operation": "search",
"time": "226.265",
"value": "Larry Ross"
},
{
"operation": "Now on",
"time": "227.251",
"value": "Larry Ross"
},
{
"operation": "Highlighting",
"time": "342.077",
"value": "Larry Ross_sentence_6"
},
{
"operation": "Highlighting",
"time": "443.131",
"value": "Larry Ross_sentence_7"
},
{
"operation": "Highlighting",
"time": "464.755",
"value": "Larry Ross_sentence_8"
},
{
"operation": "Highlighting deleted",
"time": "834.482",
"value": "Larry Ross_sentence_8"
},
{
"operation": "Highlighting deleted",
"time": "838.858",
"value": "Larry Ross_sentence_7"
},
{
"operation": "finish",
"time": "980.506",
"value": "finish"
}
]
}
|
77922
|
Urethrocele is the prolapse of the female urethra into the ovary, often occur with cystoceles but has no term for that.
|
{
"evidence": [
"Urethrocele_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:\nUrethrocele is the prolapse of the female urethra into the ovary, often occur with cystoceles but has no term for that.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Urethrocele_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1744,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.889",
"value": "?search="
},
{
"operation": "search",
"time": "128.334",
"value": "Urethrocele"
},
{
"operation": "Now on",
"time": "129.283",
"value": "Urethrocele"
},
{
"operation": "Highlighting",
"time": "135.857",
"value": "Urethrocele_sentence_1"
},
{
"operation": "Highlighting",
"time": "141.933",
"value": "Urethrocele_sentence_3"
},
{
"operation": "Highlighting",
"time": "151.739",
"value": "Urethrocele_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "153.184",
"value": "Urethrocele_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "156.686",
"value": "Urethrocele_sentence_1"
},
{
"operation": "finish",
"time": "411.582",
"value": "finish"
}
]
}
|
exfever_7988
|
Gerald Ford served as the leader of a contemporary political party in the United States that was simultaneously strengthened by the collapse of the political party that espoused traditionalist conservatism during the middle of the 19th century.
|
{
"evidence": [
"Gerald Ford",
"Republican Party (United States)",
"Whig Party (United States)"
],
"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:\nGerald Ford served as the leader of a contemporary political party in the United States that was simultaneously strengthened by the collapse of the political party that espoused traditionalist conservatism during the middle of the 19th century.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Gerald Ford",
"Republican Party (United States)",
"Whig Party (United States)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1745,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
73314
|
Jeillo Edwards who began performing at the age of four was born in Freetown, Sierra Leone and studied at the Guildhall School of Music and Drama.
|
{
"evidence": [
"Jeillo Edwards_sentence_5",
"Jeillo Edwards_sentence_3",
"Jeillo Edwards_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:\nJeillo Edwards who began performing at the age of four was born in Freetown, Sierra Leone and studied at the Guildhall School of Music and Drama.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Jeillo Edwards_sentence_5",
"Jeillo Edwards_sentence_3",
"Jeillo Edwards_sentence_4"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1746,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.137",
"value": "?search="
},
{
"operation": "search",
"time": "11.09",
"value": "Jeillo Edwards"
},
{
"operation": "Now on",
"time": "12.649",
"value": "Jeillo Edwards"
},
{
"operation": "Highlighting",
"time": "236.88",
"value": "Jeillo Edwards_sentence_5"
},
{
"operation": "Highlighting",
"time": "240.051",
"value": "Jeillo Edwards_sentence_3"
},
{
"operation": "Highlighting",
"time": "242.891",
"value": "Jeillo Edwards_sentence_4"
},
{
"operation": "finish",
"time": "256.617",
"value": "finish"
}
]
}
|
85569
|
Maureen Bruno Roy, a United States cyclocross and mountain bike racer, has competed in two UCI Cyclocross World Championships on the US National Team, and has been a four time US National Cyclocross champion in both the single speed and masters division.
|
{
"evidence": [
"Maureen Bruno Roy_sentence_0",
"Maureen Bruno Roy_sentence_1",
"Maureen Bruno Roy_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:\nMaureen Bruno Roy, a United States cyclocross and mountain bike racer, has competed in two UCI Cyclocross World Championships on the US National Team, and has been a four time US National Cyclocross champion in both the single speed and masters division.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Maureen Bruno Roy_sentence_0",
"Maureen Bruno Roy_sentence_1",
"Maureen Bruno Roy_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1747,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.113",
"value": "?search="
},
{
"operation": "search",
"time": "9.047",
"value": "Maureen Bruno Roy"
},
{
"operation": "Now on",
"time": "9.775",
"value": "Maureen Bruno Roy"
},
{
"operation": "Highlighting",
"time": "84.725",
"value": "Maureen Bruno Roy_sentence_0"
},
{
"operation": "Highlighting",
"time": "180.131",
"value": "Maureen Bruno Roy_sentence_1"
},
{
"operation": "Highlighting",
"time": "186.432",
"value": "Maureen Bruno Roy_sentence_2"
},
{
"operation": "finish",
"time": "196.094",
"value": "finish"
}
]
}
|
86389
|
Sphedanolestes cingulatus are almost completely black, the connexivum is black with red stripes but has weak femurs.
|
{
"evidence": [
"Sphedanolestes cingulatus_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:\nSphedanolestes cingulatus are almost completely black, the connexivum is black with red stripes but has weak femurs.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Sphedanolestes cingulatus_sentence_8"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1748,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.188",
"value": "?search="
},
{
"operation": "search",
"time": "10.42",
"value": "Sphedanolestes cingulatus"
},
{
"operation": "Now on",
"time": "11.327",
"value": "Sphedanolestes cingulatus"
},
{
"operation": "Highlighting",
"time": "18.91",
"value": "Sphedanolestes cingulatus_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "62.471",
"value": "Sphedanolestes cingulatus_sentence_0"
},
{
"operation": "Highlighting",
"time": "65.578",
"value": "Sphedanolestes cingulatus_sentence_6"
},
{
"operation": "Highlighting",
"time": "73.835",
"value": "Sphedanolestes cingulatus_sentence_7"
},
{
"operation": "Highlighting",
"time": "86.161",
"value": "Sphedanolestes cingulatus_sentence_8"
},
{
"operation": "Highlighting deleted",
"time": "89.04",
"value": "Sphedanolestes cingulatus_sentence_7"
},
{
"operation": "Highlighting deleted",
"time": "89.892",
"value": "Sphedanolestes cingulatus_sentence_6"
},
{
"operation": "finish",
"time": "147.537",
"value": "finish"
}
]
}
|
exfever_5133
|
Tilak was the youngest independence activist.
|
{
"evidence": [
"Bal Gangadhar Tilak",
"Lal Bal Pal"
],
"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:\nTilak was the youngest independence activist.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Bal Gangadhar Tilak",
"Lal Bal Pal"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1749,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
24926
|
Mary Melissa's father purchased a house for her in Brompton Square, but she moved to the more fashionable Eaton Square but those were her only listed residence.
|
{
"evidence": [
"Mary Melissa Hoadley Dodge_sentence_7"
],
"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:\nMary Melissa's father purchased a house for her in Brompton Square, but she moved to the more fashionable Eaton Square but those were her only listed residence.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Mary Melissa Hoadley Dodge_sentence_7"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1750,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "3.053",
"value": "?search="
},
{
"operation": "search",
"time": "14.689",
"value": "Mary Melissa Hoadley Dodge"
},
{
"operation": "Now on",
"time": "16.132",
"value": "Mary Melissa Hoadley Dodge"
},
{
"operation": "Highlighting",
"time": "30.233",
"value": "Mary Melissa Hoadley Dodge_sentence_5"
},
{
"operation": "Highlighting deleted",
"time": "50.884",
"value": "Mary Melissa Hoadley Dodge_sentence_5"
},
{
"operation": "Highlighting",
"time": "52.532",
"value": "Mary Melissa Hoadley Dodge_sentence_7"
},
{
"operation": "finish",
"time": "70.418",
"value": "finish"
}
]
}
|
20709
|
Mario Robinson is a self-taught painter, pastellist, and draftsman.
|
{
"evidence": [
"Mario Robinson_sentence_3",
"Pratt Institute_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:\nMario Robinson is a self-taught painter, pastellist, and draftsman.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Mario Robinson_sentence_3",
"Pratt Institute_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1751,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.641",
"value": "?search="
},
{
"operation": "search",
"time": "6.53",
"value": "Mario Robinson"
},
{
"operation": "Now on",
"time": "7.403",
"value": "Mario Robinson"
},
{
"operation": "Highlighting",
"time": "42.492",
"value": "Mario Robinson_sentence_3"
},
{
"operation": "hyperlink",
"time": "44.222",
"value": "Pratt Institute"
},
{
"operation": "Now on",
"time": "46.344",
"value": "Pratt Institute"
},
{
"operation": "Highlighting",
"time": "51.414",
"value": "Pratt Institute_sentence_2"
},
{
"operation": "finish",
"time": "58.687",
"value": "finish"
}
]
}
|
exfever_5707
|
Lyndon B. Johnson was an American politician who served as the 40th president of the United States known for being a member of the Democrat from a state in the South Central region of the United States.
|
{
"evidence": [
"Lyndon B. Johnson",
"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:\nLyndon B. Johnson was an American politician who served as the 40th president of the United States known for being a member of the Democrat from a state in the South Central region of the United States.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Lyndon B. Johnson",
"Texas"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1752,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_42
|
Ghost in the Shell is a cyberpunk-themed third-person shooter video game developed for the PlayStation by Exact and an American animation studio and production company formed on December 15, 1987, by Mitsuhisa Ishikawa.
|
{
"evidence": [
"Ghost in the Shell (video game)",
"Production I.G"
],
"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:\nGhost in the Shell is a cyberpunk-themed third-person shooter video game developed for the PlayStation by Exact and an American animation studio and production company formed on December 15, 1987, by Mitsuhisa Ishikawa.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Ghost in the Shell (video game)",
"Production I.G"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1753,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
87965
|
A key character in a series of children's picture books written and illustrated by Mairi Hedderwick is "Grannie Island", who lives further round the bay and was widely hailed, as for example "a positive image, a celebration of the strength of women, and a challenge to gender stereotyping".
|
{
"evidence": [
"Katie Morag_sentence_8",
"Katie Morag_sentence_2",
"Katie Morag_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:\nA key character in a series of children's picture books written and illustrated by Mairi Hedderwick is \"Grannie Island\", who lives further round the bay and was widely hailed, as for example \"a positive image, a celebration of the strength of women, and a challenge to gender stereotyping\".\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Katie Morag_sentence_8",
"Katie Morag_sentence_2",
"Katie Morag_sentence_9"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1754,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.251",
"value": "?search="
},
{
"operation": "search",
"time": "17.705",
"value": "Mairi Hedderwick"
},
{
"operation": "Now on",
"time": "19.086",
"value": "Mairi Hedderwick"
},
{
"operation": "Page search",
"time": "32.19",
"value": "Island"
},
{
"operation": "Page search",
"time": "54.819",
"value": "character"
},
{
"operation": "Page search",
"time": "88.517",
"value": "further round the bay"
},
{
"operation": "search",
"time": "105.295",
"value": "Contains...Grannie Island"
},
{
"operation": "Now on",
"time": "106.912",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "120.443",
"value": "Katie Morag"
},
{
"operation": "Now on",
"time": "121.186",
"value": "Katie Morag"
},
{
"operation": "Page search",
"time": "133.22",
"value": "Grannie Island"
},
{
"operation": "Highlighting",
"time": "135.618",
"value": "Katie Morag_sentence_8"
},
{
"operation": "Highlighting",
"time": "168.015",
"value": "Katie Morag_sentence_2"
},
{
"operation": "Highlighting",
"time": "204.881",
"value": "Katie Morag_sentence_9"
},
{
"operation": "finish",
"time": "234.802",
"value": "finish"
}
]
}
|
exfever_5428
|
"Dirty Diana" is the ninth track on Jackson's seventh studio album released on August 31, 1987, by a Canadian record label owned by Sony Music Entertainment.
|
{
"evidence": [
"Dirty Diana",
"Bad (album)",
"Epic Records"
],
"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\"Dirty Diana\" is the ninth track on Jackson's seventh studio album released on August 31, 1987, by a Canadian record label owned by Sony Music Entertainment.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Dirty Diana",
"Bad (album)",
"Epic Records"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1755,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
48540
|
Quartz fiber is mainly used for producing composite materials in electrical applications and can also be used for physical devices.
|
{
"evidence": [
"Quartz fiber_sentence_10",
"Quartz fiber_sentence_12"
],
"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:\nQuartz fiber is mainly used for producing composite materials in electrical applications and can also be used for physical devices.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Quartz fiber_sentence_10",
"Quartz fiber_sentence_12"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1756,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.996",
"value": "?search="
},
{
"operation": "search",
"time": "7.623",
"value": "Quartz fiber"
},
{
"operation": "Now on",
"time": "8.334",
"value": "Quartz fiber"
},
{
"operation": "Page search",
"time": "13.754",
"value": "compos"
},
{
"operation": "Page search",
"time": "34.297",
"value": "compo"
},
{
"operation": "Highlighting",
"time": "38.482",
"value": "Quartz fiber_sentence_10"
},
{
"operation": "Highlighting",
"time": "56.744",
"value": "Quartz fiber_sentence_12"
},
{
"operation": "finish",
"time": "120.929",
"value": "finish"
}
]
}
|
exfever_5480
|
Guillermo del Toro's eighth directorial film is Pan's Labyrinth.
|
{
"evidence": [
"Guillermo del Toro",
"Cronos (film)",
"Academy Award for Best Foreign Language 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:\nGuillermo del Toro's eighth directorial film is Pan's Labyrinth.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Guillermo del Toro",
"Cronos (film)",
"Academy Award for Best Foreign Language Film"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1757,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
59481
|
Tales of the Black Widowers' first British edition was issued by Panther, it has been translated into German and is the third of ten that describe mysteries solved by the Black Widower.
|
{
"evidence": [
"Tales of the Black Widowers_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:\nTales of the Black Widowers' first British edition was issued by Panther, it has been translated into German and is the third of ten that describe mysteries solved by the Black Widower.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Tales of the Black Widowers_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1758,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.142",
"value": "?search="
},
{
"operation": "search",
"time": "13.217",
"value": "Tales of the Black Widowers"
},
{
"operation": "Now on",
"time": "13.828",
"value": "Tales of the Black Widowers"
},
{
"operation": "Highlighting",
"time": "19.363",
"value": "Tales of the Black Widowers_sentence_2"
},
{
"operation": "Highlighting",
"time": "22.155",
"value": "Tales of the Black Widowers_sentence_3"
},
{
"operation": "Highlighting",
"time": "31.3",
"value": "Tales of the Black Widowers_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "49.199",
"value": "Tales of the Black Widowers_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "49.715",
"value": "Tales of the Black Widowers_sentence_2"
},
{
"operation": "finish",
"time": "63.538",
"value": "finish"
}
]
}
|
93808
|
Radio Continental, an Argentine News and Talk radio station owned by Grupo Latino de Radio, is founded on September 28, 1969 under the same name it has in the present.
|
{
"evidence": [
"Radio Continental_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:\nRadio Continental, an Argentine News and Talk radio station owned by Grupo Latino de Radio, is founded on September 28, 1969 under the same name it has in the present.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Radio Continental_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1759,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.085",
"value": "?search="
},
{
"operation": "search",
"time": "170.269",
"value": "Radio Continental"
},
{
"operation": "Now on",
"time": "171.271",
"value": "Radio Continental"
},
{
"operation": "Highlighting",
"time": "176.5",
"value": "Radio Continental_sentence_0"
},
{
"operation": "Highlighting",
"time": "183.093",
"value": "Radio Continental_sentence_4"
},
{
"operation": "Highlighting",
"time": "191.578",
"value": "Radio Continental_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "194.315",
"value": "Radio Continental_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "194.871",
"value": "Radio Continental_sentence_0"
},
{
"operation": "finish",
"time": "202.405",
"value": "finish"
}
]
}
|
exfever_2813
|
Chris McKenna is a French television writer who has written for an American television sitcom starring an American actress and director.
|
{
"evidence": [
"Chris McKenna (writer)",
"Community (TV series)",
"Gillian Jacobs"
],
"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:\nChris McKenna is a French television writer who has written for an American television sitcom starring an American actress and director.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Chris McKenna (writer)",
"Community (TV series)",
"Gillian Jacobs"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1760,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
36675
|
After Aerosmith abandoned the song, Jacks and his wife recorded "Seasons in the Sun" in 1973 which he released on his own label, Goldfish Records.
|
{
"evidence": [
"Seasons in the Sun_sentence_12",
"Seasons in the Sun_sentence_14"
],
"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:\nAfter Aerosmith abandoned the song, Jacks and his wife recorded \"Seasons in the Sun\" in 1973 which he released on his own label, Goldfish Records.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Seasons in the Sun_sentence_12",
"Seasons in the Sun_sentence_14"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1761,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.391",
"value": "?search="
},
{
"operation": "search",
"time": "26.608",
"value": "Seasons in the Sun"
},
{
"operation": "Now on",
"time": "29.717",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "33.043",
"value": "Seasons in the Sun"
},
{
"operation": "Now on",
"time": "33.83",
"value": "Seasons in the Sun"
},
{
"operation": "Highlighting",
"time": "58.553",
"value": "Seasons in the Sun_sentence_2"
},
{
"operation": "Page search",
"time": "65.161",
"value": "aero"
},
{
"operation": "Page search",
"time": "65.312",
"value": "aero"
},
{
"operation": "Page search",
"time": "65.447",
"value": "aero"
},
{
"operation": "Page search",
"time": "81.474",
"value": "wife"
},
{
"operation": "Page search",
"time": "81.64",
"value": "wife"
},
{
"operation": "Page search",
"time": "81.832",
"value": "wife"
},
{
"operation": "Page search",
"time": "81.993",
"value": "wife"
},
{
"operation": "Highlighting",
"time": "111.109",
"value": "Seasons in the Sun_sentence_12"
},
{
"operation": "Highlighting deleted",
"time": "195.98",
"value": "Seasons in the Sun_sentence_2"
},
{
"operation": "Page search",
"time": "215.736",
"value": "label"
},
{
"operation": "Page search",
"time": "215.943",
"value": "label"
},
{
"operation": "Page search",
"time": "216.159",
"value": "label"
},
{
"operation": "Highlighting",
"time": "225.213",
"value": "Seasons in the Sun_sentence_14"
},
{
"operation": "Highlighting",
"time": "251.106",
"value": "Seasons in the Sun_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "260.689",
"value": "Seasons in the Sun_sentence_1"
},
{
"operation": "Page search",
"time": "349.907",
"value": "aerosmith"
},
{
"operation": "Page search",
"time": "350.132",
"value": "aerosmith"
},
{
"operation": "Page search",
"time": "350.329",
"value": "aerosmith"
},
{
"operation": "hyperlink",
"time": "357.404",
"value": "Jacques Brel"
},
{
"operation": "Now on",
"time": "358.881",
"value": "Jacques Brel"
},
{
"operation": "Page search",
"time": "365.767",
"value": "aerosmith"
},
{
"operation": "Page search",
"time": "365.963",
"value": "aerosmith"
},
{
"operation": "Page search",
"time": "366.17",
"value": "aerosmith"
},
{
"operation": "Page search",
"time": "366.378",
"value": "aerosmith"
},
{
"operation": "Page search",
"time": "366.53",
"value": "aerosmith"
},
{
"operation": "back-button-clicked",
"time": "406.894",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "407.422",
"value": "Seasons in the Sun"
},
{
"operation": "finish",
"time": "418.864",
"value": "finish"
}
]
}
|
42268
|
Donauinselfest, a yearly music festival in Portugal that takes place every July in Sines, in the Alentejo region of Portugal, since 1999, has received abundant praise from the press and the artists alike and was considered "one of the best festivals in Europe" on Finnish accordion player Kimmo Pohjonen's website in 2005.
|
{
"evidence": [
"Excuse Me Moses_sentence_14"
],
"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:\nDonauinselfest, a yearly music festival in Portugal that takes place every July in Sines, in the Alentejo region of Portugal, since 1999, has received abundant praise from the press and the artists alike and was considered \"one of the best festivals in Europe\" on Finnish accordion player Kimmo Pohjonen's website in 2005.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Excuse Me Moses_sentence_14"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1762,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.184",
"value": "?search="
},
{
"operation": "search",
"time": "12.096",
"value": "Contains...Donauinselfest"
},
{
"operation": "Now on",
"time": "12.676",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "25.674",
"value": "Hitradio Ö3"
},
{
"operation": "Now on",
"time": "26.441",
"value": "Hitradio Ö3"
},
{
"operation": "Highlighting",
"time": "42.522",
"value": "Hitradio Ö3_sentence_15"
},
{
"operation": "Highlighting",
"time": "42.523",
"value": "Hitradio Ö3_title"
},
{
"operation": "Highlighting",
"time": "42.523",
"value": "Hitradio Ö3_section_1"
},
{
"operation": "Highlighting",
"time": "61.723",
"value": "Hitradio Ö3_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "75.399",
"value": "Hitradio Ö3_sentence_15"
},
{
"operation": "Highlighting deleted",
"time": "75.401",
"value": "Hitradio Ö3_section_1"
},
{
"operation": "search",
"time": "86.209",
"value": "Contains...Donauinselfest music festival"
},
{
"operation": "Now on",
"time": "87.411",
"value": "undefined"
},
{
"operation": "search",
"time": "114.04",
"value": "Sines"
},
{
"operation": "Now on",
"time": "114.898",
"value": "Sines"
},
{
"operation": "search",
"time": "141.795",
"value": "Kimmo Pohjonen"
},
{
"operation": "Now on",
"time": "142.529",
"value": "Kimmo Pohjonen"
},
{
"operation": "search",
"time": "189.621",
"value": "Contains...Donauinselfest"
},
{
"operation": "Now on",
"time": "190.217",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "198.455",
"value": "Excuse Me Moses"
},
{
"operation": "Now on",
"time": "198.974",
"value": "Excuse Me Moses"
},
{
"operation": "Highlighting",
"time": "207.747",
"value": "Excuse Me Moses_sentence_14"
},
{
"operation": "back-button-clicked",
"time": "218.464",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "219.061",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "220.568",
"value": "Hitradio Ö3"
},
{
"operation": "Now on",
"time": "220.967",
"value": "Hitradio Ö3"
},
{
"operation": "Highlighting",
"time": "228.059",
"value": "Hitradio Ö3_sentence_15"
},
{
"operation": "Highlighting",
"time": "228.06",
"value": "Hitradio Ö3_section_1"
},
{
"operation": "Highlighting deleted",
"time": "238.08",
"value": "Hitradio Ö3_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "240.981",
"value": "Hitradio Ö3_sentence_15"
},
{
"operation": "Highlighting deleted",
"time": "240.983",
"value": "Hitradio Ö3_title"
},
{
"operation": "Highlighting deleted",
"time": "240.985",
"value": "Hitradio Ö3_section_1"
},
{
"operation": "finish",
"time": "256.581",
"value": "finish"
}
]
}
|
69145
|
Flores Chaviano, considered as one of the most important Cuban composers, has received commissions from CDMC, the Fundación Príncipe de Asturias, the Radio Nacional de España and the Conservatorio de Nalón and was the president of the composers organization.
|
{
"evidence": [
"Flores Chaviano_sentence_6",
"Flores Chaviano_sentence_7",
"Flores Chaviano_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:\nFlores Chaviano, considered as one of the most important Cuban composers, has received commissions from CDMC, the Fundación Príncipe de Asturias, the Radio Nacional de España and the Conservatorio de Nalón and was the president of the composers organization.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Flores Chaviano_sentence_6",
"Flores Chaviano_sentence_7",
"Flores Chaviano_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1763,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.597",
"value": "?search="
},
{
"operation": "search",
"time": "20.548",
"value": "Flores Chaviano"
},
{
"operation": "Now on",
"time": "21.422",
"value": "Flores Chaviano"
},
{
"operation": "Highlighting",
"time": "31.459",
"value": "Flores Chaviano_sentence_0"
},
{
"operation": "Page search",
"time": "38.001",
"value": "cdmc"
},
{
"operation": "Highlighting",
"time": "43.361",
"value": "Flores Chaviano_sentence_6"
},
{
"operation": "Highlighting",
"time": "47.701",
"value": "Flores Chaviano_sentence_7"
},
{
"operation": "Highlighting",
"time": "49.648",
"value": "Flores Chaviano_sentence_5"
},
{
"operation": "Highlighting deleted",
"time": "51.495",
"value": "Flores Chaviano_sentence_0"
},
{
"operation": "finish",
"time": "59.131",
"value": "finish"
}
]
}
|
88532
|
Borys Buryak is a member of the Union of Artists of Ukraine, whose works depict Ukrainian landscapes and urban views, or draw on the traditions of Ukrainian iconography, a branch of art history which studies the identification, description, and the interpretation of the content of images.
|
{
"evidence": [
"Borys Buryak_sentence_3",
"Borys Buryak_sentence_1",
"Iconography_sentence_3"
],
"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:\nBorys Buryak is a member of the Union of Artists of Ukraine, whose works depict Ukrainian landscapes and urban views, or draw on the traditions of Ukrainian iconography, a branch of art history which studies the identification, description, and the interpretation of the content of images.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Borys Buryak_sentence_3",
"Borys Buryak_sentence_1",
"Iconography_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1764,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.228",
"value": "?search="
},
{
"operation": "search",
"time": "42.884",
"value": "Borys Buryak"
},
{
"operation": "Now on",
"time": "43.793",
"value": "Borys Buryak"
},
{
"operation": "Highlighting",
"time": "107.659",
"value": "Borys Buryak_sentence_3"
},
{
"operation": "Highlighting",
"time": "116.617",
"value": "Borys Buryak_sentence_1"
},
{
"operation": "hyperlink",
"time": "128.873",
"value": "iconography"
},
{
"operation": "Now on",
"time": "129.765",
"value": "Iconography"
},
{
"operation": "Highlighting",
"time": "134.689",
"value": "Iconography_sentence_3"
},
{
"operation": "finish",
"time": "186.288",
"value": "finish"
}
]
}
|
68116
|
The 1958 Clemson Tigers football team was an American football team that represented Clemson College and was in its 19th season under head coach Frank Howard.
|
{
"evidence": [
"1958 Clemson Tigers football team_sentence_0",
"1958 Clemson Tigers football team_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:\nThe 1958 Clemson Tigers football team was an American football team that represented Clemson College and was in its 19th season under head coach Frank Howard.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"1958 Clemson Tigers football team_sentence_0",
"1958 Clemson Tigers football team_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1765,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.076",
"value": "?search="
},
{
"operation": "search",
"time": "9.511",
"value": "1958 Clemson Tigers football"
},
{
"operation": "Now on",
"time": "11.677",
"value": "1958 Clemson Tigers football team"
},
{
"operation": "Highlighting",
"time": "13.716",
"value": "1958 Clemson Tigers football team_sentence_0"
},
{
"operation": "Highlighting",
"time": "19.578",
"value": "1958 Clemson Tigers football team_sentence_1"
},
{
"operation": "finish",
"time": "24.753",
"value": "finish"
}
]
}
|
53625
|
List of American Le Mans Series circuits were mostly raced in the United States, but in 2000 they raced in Australia, from 1999-2013 they raced in different parts of Canada, in 2000 and 2001 they raced in the United Kingdom, in 2001 there was a race in Spain, and in 2000 they raced in Germany.
|
{
"evidence": [
"American Le Mans Series_sentence_16",
"American Le Mans Series_sentence_17"
],
"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:\nList of American Le Mans Series circuits were mostly raced in the United States, but in 2000 they raced in Australia, from 1999-2013 they raced in different parts of Canada, in 2000 and 2001 they raced in the United Kingdom, in 2001 there was a race in Spain, and in 2000 they raced in Germany.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"American Le Mans Series_sentence_16",
"American Le Mans Series_sentence_17"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1766,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.895",
"value": "?search="
},
{
"operation": "search",
"time": "42.856",
"value": "American Le Mans Series"
},
{
"operation": "Now on",
"time": "44.167",
"value": "American Le Mans Series"
},
{
"operation": "Highlighting",
"time": "142.861",
"value": "American Le Mans Series_sentence_2"
},
{
"operation": "Highlighting",
"time": "169.602",
"value": "American Le Mans Series_sentence_12"
},
{
"operation": "Highlighting",
"time": "187.163",
"value": "American Le Mans Series_sentence_16"
},
{
"operation": "Highlighting",
"time": "251.296",
"value": "American Le Mans Series_sentence_17"
},
{
"operation": "finish",
"time": "276.818",
"value": "finish"
}
]
}
|
exfever_5698
|
Giacomo Girolamo Casanova was an Italian adventurer whose name is written as d?a?komo d?i?r? in the language of people of the geographical region and its neighboring insular territories of a Romance field of anthropology related to the various indigenous groups that reside in the states of Europe.
|
{
"evidence": [
"Giacomo Casanova",
"Italians",
"Ethnic groups in Europe"
],
"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:\nGiacomo Girolamo Casanova was an Italian adventurer whose name is written as d?a?komo d?i?r? in the language of people of the geographical region and its neighboring insular territories of a Romance field of anthropology related to the various indigenous groups that reside in the states of Europe.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Giacomo Casanova",
"Italians",
"Ethnic groups in Europe"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1767,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7996
|
Star vs. the Forces of Evil was nominated for Emmy Award for Outstanding Voice-Over Performance.
|
{
"evidence": [
"Star vs. the Forces of Evil",
"Disney Television Animation"
],
"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:\nStar vs. the Forces of Evil was nominated for Emmy Award for Outstanding Voice-Over Performance.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Star vs. the Forces of Evil",
"Disney Television Animation"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1768,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4547
|
How to Train Your Dragon is a 2010 American motion-captured action fantasy film produced by an American animation studio that produces animated films and television programs and is a subsidiary of Universal Pictures.
|
{
"evidence": [
"How to Train Your Dragon (film)",
"DreamWorks Animation"
],
"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:\nHow to Train Your Dragon is a 2010 American motion-captured action fantasy film produced by an American animation studio that produces animated films and television programs and is a subsidiary of Universal Pictures.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"How to Train Your Dragon (film)",
"DreamWorks Animation"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1769,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
26953
|
George Brewer (born 1766) was an English miscellaneous writer.
|
{
"evidence": [
"George Brewer_sentence_3"
],
"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:\nGeorge Brewer (born 1766) was an English miscellaneous writer.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"George Brewer_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1770,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.392",
"value": "?search="
},
{
"operation": "search",
"time": "45.569",
"value": "George Brewer"
},
{
"operation": "Now on",
"time": "46.254",
"value": "George Brewer"
},
{
"operation": "Highlighting",
"time": "50.187",
"value": "George Brewer_sentence_3"
},
{
"operation": "finish",
"time": "57.974",
"value": "finish"
}
]
}
|
exfever_1829
|
Victor-Marie Hugo played Cromwell and drama Hernani at the forefront of the artistic, literary, musical, and intellectual movement that originated in Asia.
|
{
"evidence": [
"Victor Hugo",
"Romanticism"
],
"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:\nVictor-Marie Hugo played Cromwell and drama Hernani at the forefront of the artistic, literary, musical, and intellectual movement that originated in Asia.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Victor Hugo",
"Romanticism"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1771,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
12706
|
Mosquito (Spanish and Portuguese for "little fly") control has three toxicants one of which is DDT.
|
{
"evidence": [
"Mosquito_sentence_4",
"DDT_sentence_14"
],
"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:\nMosquito (Spanish and Portuguese for \"little fly\") control has three toxicants one of which is DDT.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Mosquito_sentence_4",
"DDT_sentence_14"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1772,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.558",
"value": "?search="
},
{
"operation": "search",
"time": "4.258",
"value": "Mosquito"
},
{
"operation": "Now on",
"time": "5.372",
"value": "Mosquito"
},
{
"operation": "Highlighting",
"time": "10.377",
"value": "Mosquito_sentence_4"
},
{
"operation": "Page search",
"time": "23.738",
"value": "toxicants"
},
{
"operation": "Page search",
"time": "32.832",
"value": "toxicant"
},
{
"operation": "Page search",
"time": "46.985",
"value": "DDT"
},
{
"operation": "search",
"time": "83.85",
"value": "DDT"
},
{
"operation": "Now on",
"time": "84.912",
"value": "DDT"
},
{
"operation": "Page search",
"time": "92.077",
"value": "Mosquito"
},
{
"operation": "Highlighting",
"time": "104.573",
"value": "DDT_sentence_14"
},
{
"operation": "hyperlink",
"time": "163.826",
"value": "National Toxicology Program"
},
{
"operation": "Now on",
"time": "164.507",
"value": "National Toxicology Program"
},
{
"operation": "Page search",
"time": "169.237",
"value": "mosq"
},
{
"operation": "finish",
"time": "177.285",
"value": "finish"
}
]
}
|
26285
|
Australian World War One flying ace, George Lingham, scored six aerial victories in 1918, destroying two enemy fighters and driving four planes down, but he was never awarded for his actions.
|
{
"evidence": [
"George Lingham_sentence_7"
],
"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:\nAustralian World War One flying ace, George Lingham, scored six aerial victories in 1918, destroying two enemy fighters and driving four planes down, but he was never awarded for his actions.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"George Lingham_sentence_7"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1773,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.168",
"value": "?search="
},
{
"operation": "search",
"time": "9.998",
"value": "George Lingham"
},
{
"operation": "Now on",
"time": "10.697",
"value": "George Lingham"
},
{
"operation": "Highlighting",
"time": "41.006",
"value": "George Lingham_sentence_0"
},
{
"operation": "hyperlink",
"time": "54.029",
"value": "Royal Flying Corps"
},
{
"operation": "Now on",
"time": "55.008",
"value": "Royal Flying Corps"
},
{
"operation": "back-button-clicked",
"time": "69.319",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "69.397",
"value": "George Lingham"
},
{
"operation": "Highlighting",
"time": "80.502",
"value": "George Lingham_sentence_7"
},
{
"operation": "finish",
"time": "96.629",
"value": "finish"
}
]
}
|
63287
|
The daughter of Audebert de Noves, Laura, was a great influence on humanist poet Francesco Petrarch's works.
|
{
"evidence": [
"Laura de Noves_sentence_2",
"Laura de Noves_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:\nThe daughter of Audebert de Noves, Laura, was a great influence on humanist poet Francesco Petrarch's works.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Laura de Noves_sentence_2",
"Laura de Noves_sentence_5"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 1774,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.9",
"value": "?search="
},
{
"operation": "search",
"time": "17.525",
"value": "Contains...Audebert de Noves"
},
{
"operation": "Now on",
"time": "19.364",
"value": "undefined"
},
{
"operation": "search",
"time": "64.259",
"value": "Laura de Noves"
},
{
"operation": "Now on",
"time": "65.44",
"value": "Laura de Noves"
},
{
"operation": "Highlighting",
"time": "82.869",
"value": "Laura de Noves_sentence_2"
},
{
"operation": "Highlighting",
"time": "118.47",
"value": "Laura de Noves_sentence_5"
},
{
"operation": "finish",
"time": "176.997",
"value": "finish"
}
]
}
|
exfever_1228
|
Robert John Wagner Jr. known for starring in the television shows like an Canadian mystery television series that premiered on August 25, 1979, on ABC.
|
{
"evidence": [
"Robert Wagner",
"Hart to Hart"
],
"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:\nRobert John Wagner Jr. known for starring in the television shows like an Canadian mystery television series that premiered on August 25, 1979, on ABC.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Robert Wagner",
"Hart to Hart"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1775,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7485
|
Cry Freedom is a 1988 epic apartheid drama film directed and produced by Richard Attenborough that centres on real-life events involving a South African anti-apartheid activist who was at the forefront of a grassroots anti-apartheid campaign.
|
{
"evidence": [
"Cry Freedom",
"Steve Biko"
],
"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:\nCry Freedom is a 1988 epic apartheid drama film directed and produced by Richard Attenborough that centres on real-life events involving a South African anti-apartheid activist who was at the forefront of a grassroots anti-apartheid campaign.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Cry Freedom",
"Steve Biko"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1776,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
52477
|
Rapper Saweetie has released 17 music videos, eight promotional singles, and 20 different singles including Icy Girl that went platinum and My Type that went double platinum.
|
{
"evidence": [
"Saweetie discography_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:\nRapper Saweetie has released 17 music videos, eight promotional singles, and 20 different singles including Icy Girl that went platinum and My Type that went double platinum.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Saweetie discography_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1777,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.671",
"value": "?search="
},
{
"operation": "search",
"time": "21.35",
"value": "Rapper Saweetie"
},
{
"operation": "Now on",
"time": "22.189",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "25.214",
"value": "Saweetie"
},
{
"operation": "Now on",
"time": "25.789",
"value": "Saweetie"
},
{
"operation": "Highlighting",
"time": "29.285",
"value": "Saweetie_sentence_0"
},
{
"operation": "Highlighting",
"time": "29.285",
"value": "Saweetie_title"
},
{
"operation": "Page search",
"time": "79.335",
"value": "icy"
},
{
"operation": "Page search",
"time": "125.384",
"value": "platinum"
},
{
"operation": "Highlighting",
"time": "131.496",
"value": "Saweetie_sentence_22"
},
{
"operation": "Highlighting",
"time": "131.496",
"value": "Saweetie_section_1"
},
{
"operation": "Highlighting",
"time": "131.496",
"value": "Saweetie_section_2"
},
{
"operation": "Highlighting",
"time": "132.624",
"value": "Saweetie_sentence_23"
},
{
"operation": "Highlighting",
"time": "141.88",
"value": "Saweetie_sentence_29"
},
{
"operation": "Highlighting",
"time": "141.881",
"value": "Saweetie_section_3"
},
{
"operation": "Page search",
"time": "151.608",
"value": "videos"
},
{
"operation": "Page search",
"time": "170.144",
"value": "music"
},
{
"operation": "Page search",
"time": "190.449",
"value": "singles"
},
{
"operation": "hyperlink",
"time": "264.843",
"value": "Saweetie discography"
},
{
"operation": "Now on",
"time": "265.499",
"value": "Saweetie discography"
},
{
"operation": "Highlighting",
"time": "271.123",
"value": "Saweetie discography_sentence_0"
},
{
"operation": "finish",
"time": "303.419",
"value": "finish"
}
]
}
|
exfever_1467
|
Karl Anthony Malone spent his first 18 seasons in the NBA?with the Utah Jazz and formed a formidable duo with his teammate,?regarded as one of the greatest point guards and passers of all time.
|
{
"evidence": [
"Karl Malone",
"John Stockton"
],
"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:\nKarl Anthony Malone spent his first 18 seasons in the NBA?with the Utah Jazz and formed a formidable duo with his teammate,?regarded as one of the greatest point guards and passers of all time.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Karl Malone",
"John Stockton"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 1778,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2539
|
World War I began on the 19th of July, 1914, in which the fighting expanded to the world's second-largest and second-most populous continents, after the Earth's largest and most populous continent and shares the continental landmass of Eurasia.
|
{
"evidence": [
"World War I",
"Africa",
"Asia"
],
"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:\nWorld War I began on the 19th of July, 1914, in which the fighting expanded to the world's second-largest and second-most populous continents, after the Earth's largest and most populous continent and shares the continental landmass of Eurasia.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"World War I",
"Africa",
"Asia"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1779,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3821
|
Keith Urbahn is an Australian musician with nominations to honor the best in film and television, which were broadcast by Marvel.
|
{
"evidence": [
"Keith Urban",
"70th Golden Globe Awards"
],
"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:\nKeith Urbahn is an Australian musician with nominations to honor the best in film and television, which were broadcast by Marvel.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Keith Urban",
"70th Golden Globe Awards"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1780,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2433
|
Mike Myers is a Canadian actor who conceived the premise for, and starred as Dr. Raymond Stantz in an American supernatural comedy film directed and produced by Ivan Reitman, and written by an American actor best-known film acting roles were as Egon Spengler in Ghostbusters.
|
{
"evidence": [
"Dan Aykroyd",
"Ghostbusters",
"Harold Ramis"
],
"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:\nMike Myers is a Canadian actor who conceived the premise for, and starred as Dr. Raymond Stantz in an American supernatural comedy film directed and produced by Ivan Reitman, and written by an American actor best-known film acting roles were as Egon Spengler in Ghostbusters.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Dan Aykroyd",
"Ghostbusters",
"Harold Ramis"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1781,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
40356
|
The UVA Health System's history can be traced to the first meeting of the Board of Visitors the year the University of Virginia was founded in 1918, when a School of Medicine was authorized.
|
{
"evidence": [
"University of Virginia Health System_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:\nThe UVA Health System's history can be traced to the first meeting of the Board of Visitors the year the University of Virginia was founded in 1918, when a School of Medicine was authorized.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"University of Virginia Health System_sentence_9"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1782,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.059",
"value": "?search="
},
{
"operation": "search",
"time": "104.055",
"value": "Contains...UVA Health System"
},
{
"operation": "Now on",
"time": "105.89",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "158.6",
"value": "Novant Health UVA Health System Novant Health_section_4"
},
{
"operation": "Now on",
"time": "159.409",
"value": "Novant Health#Novant Health UVA Health System Novant Health section 4"
},
{
"operation": "Page search",
"time": "172.949",
"value": "UVA health"
},
{
"operation": "search",
"time": "244.87",
"value": "University of Virginia health"
},
{
"operation": "Now on",
"time": "247.742",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "252.52",
"value": "University of Virginia Health System"
},
{
"operation": "Now on",
"time": "253.161",
"value": "University of Virginia Health System"
},
{
"operation": "Page search",
"time": "266.544",
"value": "1918"
},
{
"operation": "Highlighting",
"time": "274.996",
"value": "University of Virginia Health System_sentence_9"
},
{
"operation": "finish",
"time": "280.673",
"value": "finish"
}
]
}
|
25384
|
ShopBack received further investment funding in 2015 and was eventually raised to USD $40 million in 2016 with the help of several companies.
|
{
"evidence": [
"ShopBack_sentence_11"
],
"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:\nShopBack received further investment funding in 2015 and was eventually raised to USD $40 million in 2016 with the help of several companies.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"ShopBack_sentence_11"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1783,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.628",
"value": "?search="
},
{
"operation": "search",
"time": "13.652",
"value": "ShopBack"
},
{
"operation": "Now on",
"time": "14.541",
"value": "ShopBack"
},
{
"operation": "Page search",
"time": "29.032",
"value": "2015"
},
{
"operation": "Page search",
"time": "29.527",
"value": "2015"
},
{
"operation": "Page search",
"time": "29.87",
"value": "2015"
},
{
"operation": "Highlighting",
"time": "32.322",
"value": "ShopBack_sentence_10"
},
{
"operation": "Page search",
"time": "52.076",
"value": "2016"
},
{
"operation": "Highlighting",
"time": "78.404",
"value": "ShopBack_sentence_11"
},
{
"operation": "Highlighting deleted",
"time": "79.124",
"value": "ShopBack_sentence_10"
},
{
"operation": "finish",
"time": "86.528",
"value": "finish"
}
]
}
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.