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_4248
|
Akira Toriyama is a Japanese manga artist known for his 2007 role-playing video game developed by Mistwalker.
|
{
"evidence": [
"Akira Toriyama",
"Blue Dragon (video game)"
],
"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:\nAkira Toriyama is a Japanese manga artist known for his 2007 role-playing video game developed by Mistwalker.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Akira Toriyama",
"Blue Dragon (video game)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7584,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4353
|
British Airways is the second largest UK based carrier, behind a British multinational low-cost airline group.
|
{
"evidence": [
"British Airways",
"EasyJet"
],
"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:\nBritish Airways is the second largest UK based carrier, behind a British multinational low-cost airline group.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"British Airways",
"EasyJet"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7585,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
42791
|
Produced and directed by Marc Gracie, Surprise Surprise Gotcha pulls practical jokes on Australian celebrities.
|
{
"evidence": [
"Surprise Surprise Gotcha_sentence_5",
"Surprise Surprise Gotcha_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:\nProduced and directed by Marc Gracie, Surprise Surprise Gotcha pulls practical jokes on Australian celebrities.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Surprise Surprise Gotcha_sentence_5",
"Surprise Surprise Gotcha_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7586,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.742",
"value": "?search="
},
{
"operation": "search",
"time": "8.508",
"value": "Surprise Surprise Gotcha"
},
{
"operation": "Now on",
"time": "9.127",
"value": "Surprise Surprise Gotcha"
},
{
"operation": "Highlighting",
"time": "16.441",
"value": "Surprise Surprise Gotcha_sentence_5"
},
{
"operation": "Highlighting",
"time": "22.156",
"value": "Surprise Surprise Gotcha_sentence_3"
},
{
"operation": "finish",
"time": "27.723",
"value": "finish"
}
]
}
|
16175
|
In three consecutive years, 2004 to 2006, the United People's Freedom Alliance dominated the elections in Wellawaya but not in the Monaragala Electoral District.
|
{
"evidence": [
"Wellawaya Electoral District_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:\nIn three consecutive years, 2004 to 2006, the United People's Freedom Alliance dominated the elections in Wellawaya but not in the Monaragala Electoral District.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Wellawaya Electoral District_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7587,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.095",
"value": "?search="
},
{
"operation": "search",
"time": "14.679",
"value": "United People's Freedom Alliance"
},
{
"operation": "Now on",
"time": "15.597",
"value": "United People's Freedom Alliance"
},
{
"operation": "Page search",
"time": "73.176",
"value": "Wellawaya"
},
{
"operation": "search",
"time": "107.214",
"value": "Wellawaya"
},
{
"operation": "Now on",
"time": "108.188",
"value": "Wellawaya"
},
{
"operation": "search",
"time": "212.502",
"value": "Monaragala Electoral District"
},
{
"operation": "Now on",
"time": "214.25",
"value": "Monaragala Electoral District"
},
{
"operation": "search",
"time": "280.349",
"value": "United People's Freedom Alliance"
},
{
"operation": "Now on",
"time": "281.148",
"value": "United People's Freedom Alliance"
},
{
"operation": "Page search",
"time": "366.616",
"value": "Monaragala Electoral District"
},
{
"operation": "search",
"time": "407.845",
"value": "Monaragala Electoral District"
},
{
"operation": "Now on",
"time": "408.835",
"value": "Monaragala Electoral District"
},
{
"operation": "back-button-clicked",
"time": "414.375",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "414.478",
"value": "United People's Freedom Alliance"
},
{
"operation": "search",
"time": "422.133",
"value": "Wellawaya"
},
{
"operation": "Now on",
"time": "423.04",
"value": "Wellawaya"
},
{
"operation": "search",
"time": "440.92",
"value": "Wellawaya Electoral District"
},
{
"operation": "Now on",
"time": "441.819",
"value": "Wellawaya Electoral District"
},
{
"operation": "Highlighting",
"time": "462.968",
"value": "Wellawaya Electoral District_sentence_4"
},
{
"operation": "finish",
"time": "482.19",
"value": "finish"
}
]
}
|
exfever_5387
|
Yangon is the city that physically encompasses the government's offices and meeting places in the Yangon Region and is the largest city in Myanmar.
|
{
"evidence": [
"Yangon",
"Capital city"
],
"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:\nYangon is the city that physically encompasses the government's offices and meeting places in the Yangon Region and is the largest city in Myanmar.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Yangon",
"Capital city"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7588,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
41904
|
Melaleuca marginata is a prickly shrub with white flowers that occur mainly in early spring.
|
{
"evidence": [
"Melaleuca marginata_sentence_3",
"Melaleuca marginata_sentence_5",
"Melaleuca marginata_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:\nMelaleuca marginata is a prickly shrub with white flowers that occur mainly in early spring.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Melaleuca marginata_sentence_3",
"Melaleuca marginata_sentence_5",
"Melaleuca marginata_sentence_9"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7589,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.753",
"value": "?search="
},
{
"operation": "search",
"time": "11.812",
"value": "Melaleuca marginata"
},
{
"operation": "Now on",
"time": "12.758",
"value": "Melaleuca marginata"
},
{
"operation": "Highlighting",
"time": "19.638",
"value": "Melaleuca marginata_sentence_0"
},
{
"operation": "Highlighting",
"time": "21.279",
"value": "Melaleuca marginata_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "29.147",
"value": "Melaleuca marginata_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "29.984",
"value": "Melaleuca marginata_sentence_1"
},
{
"operation": "Highlighting",
"time": "31.757",
"value": "Melaleuca marginata_sentence_3"
},
{
"operation": "Highlighting",
"time": "36.687",
"value": "Melaleuca marginata_sentence_5"
},
{
"operation": "Highlighting",
"time": "44.201",
"value": "Melaleuca marginata_sentence_9"
},
{
"operation": "finish",
"time": "65.522",
"value": "finish"
}
]
}
|
22760
|
The Kirtlebridge rail crash took place on October 2, 1872 at Caledonian Railway, Kirtlebridge, Dumfries and Galloway, Southern Scotland, when the shunting took place without the signalman's authority.
|
{
"evidence": [
"Kirtlebridge rail crash_sentence_0",
"Kirtlebridge rail crash_sentence_5",
"Kirtlebridge rail crash_sentence_2",
"Kirtlebridge rail crash_sentence_3",
"Kirtlebridge railway station_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 Kirtlebridge rail crash took place on October 2, 1872 at Caledonian Railway, Kirtlebridge, Dumfries and Galloway, Southern Scotland, when the shunting took place without the signalman's authority.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Kirtlebridge rail crash_sentence_0",
"Kirtlebridge rail crash_sentence_5",
"Kirtlebridge rail crash_sentence_2",
"Kirtlebridge rail crash_sentence_3",
"Kirtlebridge railway station_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7590,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.562",
"value": "?search="
},
{
"operation": "search",
"time": "14.505",
"value": "Yuba City bus disaster"
},
{
"operation": "Now on",
"time": "15.401",
"value": "Yuba City bus disaster"
},
{
"operation": "back-button-clicked",
"time": "115.684",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "115.714",
"value": "?search="
},
{
"operation": "search",
"time": "123.391",
"value": "Kirtlebridge rail crash"
},
{
"operation": "Now on",
"time": "124.172",
"value": "Kirtlebridge rail crash"
},
{
"operation": "Highlighting",
"time": "149.27",
"value": "Kirtlebridge rail crash_sentence_0"
},
{
"operation": "Highlighting",
"time": "152.791",
"value": "Kirtlebridge rail crash_sentence_5"
},
{
"operation": "Highlighting",
"time": "174.752",
"value": "Kirtlebridge rail crash_sentence_2"
},
{
"operation": "Highlighting",
"time": "179.307",
"value": "Kirtlebridge rail crash_sentence_3"
},
{
"operation": "Page search",
"time": "187.268",
"value": "signal"
},
{
"operation": "Page search",
"time": "217.211",
"value": "curtlebridge"
},
{
"operation": "hyperlink",
"time": "247.66",
"value": "Caledonian Railway"
},
{
"operation": "Now on",
"time": "248.592",
"value": "Caledonian Railway"
},
{
"operation": "search",
"time": "256.26",
"value": "dumfries"
},
{
"operation": "Now on",
"time": "257.488",
"value": "Dumfries"
},
{
"operation": "back-button-clicked",
"time": "259.711",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "259.77",
"value": "Caledonian Railway"
},
{
"operation": "Page search",
"time": "263.453",
"value": "dumfries"
},
{
"operation": "back-button-clicked",
"time": "294.106",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "294.109",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "294.145",
"value": "Kirtlebridge rail crash"
},
{
"operation": "hyperlink",
"time": "302.598",
"value": "Kirtlebridge railway station"
},
{
"operation": "Now on",
"time": "303.393",
"value": "Kirtlebridge railway station"
},
{
"operation": "Highlighting",
"time": "316.697",
"value": "Kirtlebridge railway station_sentence_1"
},
{
"operation": "Page search",
"time": "321.187",
"value": "dumfries"
},
{
"operation": "Highlighting",
"time": "327.831",
"value": "Kirtlebridge railway station_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "329.794",
"value": "Kirtlebridge railway station_sentence_1"
},
{
"operation": "back-button-clicked",
"time": "334.958",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "334.988",
"value": "Kirtlebridge rail crash"
},
{
"operation": "finish",
"time": "352.839",
"value": "finish"
}
]
}
|
exfever_1672
|
Your Highness is a 2011 American stoner comic fantasy film directed by David Gordon Green that stars the American actor, comedian, producer, and writer who appeared in the 2006 American low-budget martial arts black comedy film directed by Jody Hill.
|
{
"evidence": [
"Your Highness",
"Danny McBride",
"The Foot Fist Way"
],
"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:\nYour Highness is a 2011 American stoner comic fantasy film directed by David Gordon Green that stars the American actor, comedian, producer, and writer who appeared in the 2006 American low-budget martial arts black comedy film directed by Jody Hill.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Your Highness",
"Danny McBride",
"The Foot Fist Way"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7591,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
60320
|
The oldest part of one of the oldest souks of Bab Jebli, the Souk El Hboub, which was originally called the Souk of Food, is Ahmend Bey's yard.
|
{
"evidence": [
"Souk El Hboub_sentence_0",
"Souk El Hboub_sentence_1",
"Souk El Hboub_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:\nThe oldest part of one of the oldest souks of Bab Jebli, the Souk El Hboub, which was originally called the Souk of Food, is Ahmend Bey's yard.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Souk El Hboub_sentence_0",
"Souk El Hboub_sentence_1",
"Souk El Hboub_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7592,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.471",
"value": "?search="
},
{
"operation": "search",
"time": "12.789",
"value": "Souk El Hboub"
},
{
"operation": "Now on",
"time": "13.709",
"value": "Souk El Hboub"
},
{
"operation": "Highlighting",
"time": "405.255",
"value": "Souk El Hboub_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "406.186",
"value": "Souk El Hboub_sentence_1"
},
{
"operation": "Highlighting",
"time": "457.708",
"value": "Souk El Hboub_sentence_1"
},
{
"operation": "Highlighting",
"time": "491.962",
"value": "Souk El Hboub_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "494.248",
"value": "Souk El Hboub_sentence_1"
},
{
"operation": "Highlighting",
"time": "573.536",
"value": "Souk El Hboub_sentence_1"
},
{
"operation": "Highlighting",
"time": "577.428",
"value": "Souk El Hboub_sentence_2"
},
{
"operation": "finish",
"time": "584.89",
"value": "finish"
}
]
}
|
exfever_8850
|
Ryan John Seacrest's fourth hosting is on American Idol.
|
{
"evidence": [
"Ryan Seacrest",
"Emmy Award"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nRyan John Seacrest's fourth hosting is on American Idol.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Ryan Seacrest",
"Emmy Award"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7593,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
39104
|
Conway (born on March 16, 1930) excelled in football, track and field, basketball, and baseball as a student at North Kansas City High School; then upon graduating, he was considered by many to be the best running back in football in the Kansas City.
|
{
"evidence": [
"Al Conway_sentence_0",
"Al Conway_sentence_5",
"Al Conway_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:\nConway (born on March 16, 1930) excelled in football, track and field, basketball, and baseball as a student at North Kansas City High School; then upon graduating, he was considered by many to be the best running back in football in the Kansas City.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Al Conway_sentence_0",
"Al Conway_sentence_5",
"Al Conway_sentence_4"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7594,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.658",
"value": "?search="
},
{
"operation": "search",
"time": "32.219",
"value": "Conway KANSAS CITY"
},
{
"operation": "Now on",
"time": "33.909",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "48.797",
"value": "Pete Conway"
},
{
"operation": "Now on",
"time": "49.359",
"value": "Pete Conway"
},
{
"operation": "back-button-clicked",
"time": "87.397",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "88.052",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "99.571",
"value": "next 20"
},
{
"operation": "Now on",
"time": "100.329",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "121.676",
"value": "Kansas City Cowboys (AA) all-time roster"
},
{
"operation": "Now on",
"time": "122.423",
"value": "Kansas City Cowboys (AA) all-time roster"
},
{
"operation": "hyperlink",
"time": "134.753",
"value": "Jim Conway"
},
{
"operation": "Now on",
"time": "135.258",
"value": "Jim Conway (baseball)"
},
{
"operation": "back-button-clicked",
"time": "147.38",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "147.632",
"value": "Kansas City Cowboys (AA) all-time roster"
},
{
"operation": "search",
"time": "175.117",
"value": "Contains...best running back in football"
},
{
"operation": "Now on",
"time": "177.147",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "183.088",
"value": "Al Conway"
},
{
"operation": "Now on",
"time": "183.609",
"value": "Al Conway"
},
{
"operation": "Highlighting",
"time": "188.29",
"value": "Al Conway_sentence_0"
},
{
"operation": "Highlighting",
"time": "201.233",
"value": "Al Conway_sentence_5"
},
{
"operation": "Highlighting",
"time": "228.67",
"value": "Al Conway_sentence_4"
},
{
"operation": "finish",
"time": "270.04",
"value": "finish"
}
]
}
|
16667
|
Endre Kristian Vestvik was elected to the Norwegian Parliament and died after his term ended.
|
{
"evidence": [
"Endre Kristian Vestvik_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:\nEndre Kristian Vestvik was elected to the Norwegian Parliament and died after his term ended.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Endre Kristian Vestvik_sentence_3"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7595,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.959",
"value": "?search="
},
{
"operation": "search",
"time": "13.991",
"value": "Endre Kristian Vestvik"
},
{
"operation": "Now on",
"time": "15.781",
"value": "Endre Kristian Vestvik"
},
{
"operation": "Highlighting",
"time": "43.111",
"value": "Endre Kristian Vestvik_sentence_3"
},
{
"operation": "finish",
"time": "55.735",
"value": "finish"
}
]
}
|
11452
|
San Francisco Giants won the World Series of the Major League Baseball 2010 season, winning 5-0 in five games.
|
{
"evidence": [
"2010 World Series_sentence_78",
"2010 World Series_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:\nSan Francisco Giants won the World Series of the Major League Baseball 2010 season, winning 5-0 in five games.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"2010 World Series_sentence_78",
"2010 World Series_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7596,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.9",
"value": "?search="
},
{
"operation": "search",
"time": "35.786",
"value": "2010 Major league baseball world series"
},
{
"operation": "Now on",
"time": "38.925",
"value": "undefined"
},
{
"operation": "search",
"time": "70.525",
"value": "world series of major league baseball"
},
{
"operation": "Now on",
"time": "72.535",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "78.407",
"value": "Major League Baseball postseason"
},
{
"operation": "Now on",
"time": "79.112",
"value": "Major League Baseball postseason"
},
{
"operation": "hyperlink",
"time": "88.701",
"value": "1.3 1994–2011"
},
{
"operation": "search",
"time": "114.826",
"value": "san francisco giants 2010 seaosn"
},
{
"operation": "Now on",
"time": "117.133",
"value": "undefined"
},
{
"operation": "search",
"time": "134.72",
"value": "san francisco giants 2010 season"
},
{
"operation": "Now on",
"time": "136.687",
"value": "undefined"
},
{
"operation": "search",
"time": "153.051",
"value": "world series of the major league baseball"
},
{
"operation": "Now on",
"time": "155.32",
"value": "undefined"
},
{
"operation": "search",
"time": "180.277",
"value": "2010 world series of the major league baseball"
},
{
"operation": "Now on",
"time": "181.195",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "187.083",
"value": "League organization Major League Baseball_section_1"
},
{
"operation": "Now on",
"time": "188.539",
"value": "Major League Baseball#League organization Major League Baseball section 1"
},
{
"operation": "search",
"time": "211.497",
"value": "world series of baseball"
},
{
"operation": "Now on",
"time": "212.591",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "221.571",
"value": "Before 1969"
},
{
"operation": "Now on",
"time": "221.913",
"value": "Major League Baseball postseason#Before 1969"
},
{
"operation": "hyperlink",
"time": "241.028",
"value": "1.3 1994–2011"
},
{
"operation": "search",
"time": "265.551",
"value": "San Francisco Giants"
},
{
"operation": "Now on",
"time": "267.219",
"value": "San Francisco Giants"
},
{
"operation": "Page search",
"time": "278.406",
"value": "world series"
},
{
"operation": "hyperlink",
"time": "299.928",
"value": "2010"
},
{
"operation": "Now on",
"time": "301.002",
"value": "2010 World Series"
},
{
"operation": "Highlighting",
"time": "316.577",
"value": "2010 World Series_sentence_78"
},
{
"operation": "Highlighting",
"time": "351.799",
"value": "2010 World Series_sentence_1"
},
{
"operation": "finish",
"time": "357.459",
"value": "finish"
}
]
}
|
63625
|
Rudi Strittich debuted for Vorwarts Steyr, and played for First Vienna FC.
|
{
"evidence": [
"Rudi Strittich_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:\nRudi Strittich debuted for Vorwarts Steyr, and played for First Vienna FC.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Rudi Strittich_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7597,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.956",
"value": "?search="
},
{
"operation": "search",
"time": "6.226",
"value": "Rudi Strittich"
},
{
"operation": "Now on",
"time": "7.571",
"value": "Rudi Strittich"
},
{
"operation": "Highlighting",
"time": "13.616",
"value": "Rudi Strittich_sentence_1"
},
{
"operation": "finish",
"time": "57.913",
"value": "finish"
}
]
}
|
23300
|
The 2018 U Sports University Cup Queen's Cup championship game was held in Ontario, but Alberta was penalized for an ineligible player.
|
{
"evidence": [
"2018 U Sports University Cup_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:\nThe 2018 U Sports University Cup Queen's Cup championship game was held in Ontario, but Alberta was penalized for an ineligible player.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"2018 U Sports University Cup_sentence_17"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7598,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.023",
"value": "?search="
},
{
"operation": "search",
"time": "16.808",
"value": "2018 U Sports University Cup Queen's Cup"
},
{
"operation": "Now on",
"time": "19.106",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "39.82",
"value": "David Johnston University Cup"
},
{
"operation": "Now on",
"time": "40.794",
"value": "David Johnston University Cup"
},
{
"operation": "search",
"time": "123.099",
"value": "2018 Queen's Cup"
},
{
"operation": "Now on",
"time": "124.857",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "130.129",
"value": "Miguel Tabuena"
},
{
"operation": "Now on",
"time": "130.681",
"value": "Miguel Tabuena"
},
{
"operation": "search",
"time": "155.697",
"value": "2018 U Sports University Cup"
},
{
"operation": "Now on",
"time": "156.365",
"value": "2018 U Sports University Cup"
},
{
"operation": "Highlighting",
"time": "213.745",
"value": "2018 U Sports University Cup_sentence_17"
},
{
"operation": "hyperlink",
"time": "218.704",
"value": "Queen's Cup"
},
{
"operation": "Now on",
"time": "219.466",
"value": "Queen's Cup (ice hockey)"
},
{
"operation": "back-button-clicked",
"time": "240.339",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "240.451",
"value": "2018 U Sports University Cup"
},
{
"operation": "finish",
"time": "267.72",
"value": "finish"
}
]
}
|
exfever_2106
|
Ralph Nathaniel Twisleton-Wykeham-Fiennes first achieved success as a Shakespeare interpreter onstage at one of the United Kingdom's most prominent publicly funded performing arts venues.
|
{
"evidence": [
"Ralph Fiennes",
"Royal National Theatre"
],
"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:\nRalph Nathaniel Twisleton-Wykeham-Fiennes first achieved success as a Shakespeare interpreter onstage at one of the United Kingdom's most prominent publicly funded performing arts venues.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Ralph Fiennes",
"Royal National Theatre"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7599,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
9588
|
The Turgo turbine(designed in 1919 by Gilkes) is an impulse water turbine with an 87% efficiency that operates at higher flow rates than the Pelton turbine.
|
{
"evidence": [
"Turgo turbine_sentence_4",
"Turgo turbine_sentence_0",
"Turgo turbine_sentence_1",
"Turgo turbine_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:\nThe Turgo turbine(designed in 1919 by Gilkes) is an impulse water turbine with an 87% efficiency that operates at higher flow rates than the Pelton turbine.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Turgo turbine_sentence_4",
"Turgo turbine_sentence_0",
"Turgo turbine_sentence_1",
"Turgo turbine_sentence_7"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7600,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.558",
"value": "?search="
},
{
"operation": "search",
"time": "6.795",
"value": "turgo turbine"
},
{
"operation": "Now on",
"time": "7.374",
"value": "Turgo turbine"
},
{
"operation": "Highlighting",
"time": "9.652",
"value": "Turgo turbine_sentence_4"
},
{
"operation": "Highlighting",
"time": "14.335",
"value": "Turgo turbine_sentence_0"
},
{
"operation": "Highlighting",
"time": "17.747",
"value": "Turgo turbine_sentence_1"
},
{
"operation": "Page search",
"time": "33.135",
"value": "flow"
},
{
"operation": "Highlighting",
"time": "37.016",
"value": "Turgo turbine_sentence_7"
},
{
"operation": "finish",
"time": "52.278",
"value": "finish"
}
]
}
|
71336
|
The 2018 Arkansas elections was held in the US state of Arkansas on November 6, 2018 and all of Arkansas' executive officers were up for election.
|
{
"evidence": [
"2018 Arkansas elections_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:\nThe 2018 Arkansas elections was held in the US state of Arkansas on November 6, 2018 and all of Arkansas' executive officers were up for election.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"2018 Arkansas elections_sentence_2"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7601,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.672",
"value": "?search="
},
{
"operation": "search",
"time": "9.358",
"value": "2018 Arkansas elections"
},
{
"operation": "Now on",
"time": "11.105",
"value": "2018 Arkansas elections"
},
{
"operation": "Highlighting",
"time": "33.241",
"value": "2018 Arkansas elections_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "52.506",
"value": "2018 Arkansas elections_sentence_2"
},
{
"operation": "Highlighting",
"time": "53.243",
"value": "2018 Arkansas elections_sentence_02018 Arkansas elections_sentence_1"
},
{
"operation": "Highlighting",
"time": "55.737",
"value": "2018 Arkansas elections_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "56.546",
"value": "2018 Arkansas elections_sentence_02018 Arkansas elections_sentence_1"
},
{
"operation": "finish",
"time": "62.784",
"value": "finish"
}
]
}
|
exfever_971
|
The Fate of the Furious is a 2017 Canadian action film whose sequel to a 2015 American action film directed by James Wan and written by Chris Morgan.
|
{
"evidence": [
"The Fate of the Furious",
"Furious 7"
],
"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 Fate of the Furious is a 2017 Canadian action film whose sequel to a 2015 American action film directed by James Wan and written by Chris Morgan.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Fate of the Furious",
"Furious 7"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7602,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
14113
|
Notarcha aurolinealis is a species of moth of the kingdom animalia, phylum arthropoda and class insecta which is only a small class within the arthropod phylum.
|
{
"evidence": [
"Insect_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:\nNotarcha aurolinealis is a species of moth of the kingdom animalia, phylum arthropoda and class insecta which is only a small class within the arthropod phylum.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Insect_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7603,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.583",
"value": "?search="
},
{
"operation": "search",
"time": "6.436",
"value": "notarcha aurolinealis"
},
{
"operation": "Now on",
"time": "7.654",
"value": "Notarcha aurolinealis"
},
{
"operation": "Highlighting",
"time": "8.857",
"value": "Notarcha aurolinealis_sentence_0"
},
{
"operation": "Highlighting",
"time": "8.858",
"value": "Notarcha aurolinealis_title"
},
{
"operation": "Highlighting",
"time": "11.885",
"value": "Notarcha aurolinealis_cell_0_2_1"
},
{
"operation": "Highlighting",
"time": "11.891",
"value": "Notarcha aurolinealis_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "11.891",
"value": "Notarcha aurolinealis_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "12.348",
"value": "Notarcha aurolinealis_cell_0_2_0"
},
{
"operation": "Highlighting",
"time": "14.26",
"value": "Notarcha aurolinealis_cell_0_3_1"
},
{
"operation": "Highlighting",
"time": "15.24",
"value": "Notarcha aurolinealis_cell_0_3_0"
},
{
"operation": "Highlighting",
"time": "17.185",
"value": "Notarcha aurolinealis_cell_0_5_1"
},
{
"operation": "Highlighting",
"time": "17.65",
"value": "Notarcha aurolinealis_cell_0_5_0"
},
{
"operation": "hyperlink",
"time": "20.748",
"value": "Insecta"
},
{
"operation": "Now on",
"time": "21.795",
"value": "Insect"
},
{
"operation": "Highlighting",
"time": "24.557",
"value": "Insect_sentence_1"
},
{
"operation": "Highlighting",
"time": "28.012",
"value": "Insect_sentence_1"
},
{
"operation": "finish",
"time": "36.916",
"value": "finish"
}
]
}
|
28194
|
Direct Action and Democracy Today is a 2005 book by April Carter, she is supportive of direct action and suggest that theorists have paid a lot of attention to this phenomenon.
|
{
"evidence": [
"Direct Action and Democracy Today_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:\nDirect Action and Democracy Today is a 2005 book by April Carter, she is supportive of direct action and suggest that theorists have paid a lot of attention to this phenomenon.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Direct Action and Democracy Today_sentence_3"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7604,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.973",
"value": "undefined"
},
{
"operation": "search",
"time": "7.207",
"value": ""
},
{
"operation": "Now on",
"time": "7.551",
"value": "undefined"
},
{
"operation": "search",
"time": "9.695",
"value": "Direct Action and Democracy Today"
},
{
"operation": "Now on",
"time": "10.394",
"value": "Direct Action and Democracy Today"
},
{
"operation": "Highlighting",
"time": "11.481",
"value": "Direct Action and Democracy Today_sentence_0"
},
{
"operation": "Highlighting",
"time": "16.773",
"value": "Direct Action and Democracy Today_sentence_2"
},
{
"operation": "Highlighting",
"time": "24.804",
"value": "Direct Action and Democracy Today_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "29.618",
"value": "Direct Action and Democracy Today_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "30.425",
"value": "Direct Action and Democracy Today_sentence_0"
},
{
"operation": "finish",
"time": "35.566",
"value": "finish"
}
]
}
|
66685
|
Garra regressus is in the Labeoninae (a doubtfully distinct subfamily of neither ray-finned nor lobe-finned fishes), within the Cyprinidae family, under the Cypriniformes order.
|
{
"evidence": [
"Labeoninae_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:\nGarra regressus is in the Labeoninae (a doubtfully distinct subfamily of neither ray-finned nor lobe-finned fishes), within the Cyprinidae family, under the Cypriniformes order.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Labeoninae_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7605,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.343",
"value": "?search="
},
{
"operation": "search",
"time": "9.909",
"value": "Garra regressus"
},
{
"operation": "Now on",
"time": "11.348",
"value": "Garra regressus"
},
{
"operation": "Highlighting",
"time": "16.94",
"value": "Garra regressus_cell_0_8_1"
},
{
"operation": "Highlighting",
"time": "16.941",
"value": "Garra regressus_title"
},
{
"operation": "Highlighting",
"time": "16.944",
"value": "Garra regressus_header_cell_0_2_0"
},
{
"operation": "Highlighting",
"time": "16.944",
"value": "Garra regressus_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "16.945",
"value": "Garra regressus_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "22.829",
"value": "Garra regressus_cell_0_7_1"
},
{
"operation": "Highlighting",
"time": "24.219",
"value": "Garra regressus_cell_0_7_0"
},
{
"operation": "Highlighting",
"time": "26.224",
"value": "Garra regressus_cell_0_6_1"
},
{
"operation": "Highlighting",
"time": "26.925",
"value": "Garra regressus_cell_0_6_0"
},
{
"operation": "hyperlink",
"time": "28.005",
"value": "Labeoninae"
},
{
"operation": "Now on",
"time": "28.722",
"value": "Labeoninae"
},
{
"operation": "Highlighting",
"time": "30.789",
"value": "Labeoninae_sentence_0"
},
{
"operation": "back-button-clicked",
"time": "35.973",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "36.077",
"value": "Garra regressus"
},
{
"operation": "Highlighting deleted",
"time": "37.144",
"value": "Garra regressus_cell_0_6_1"
},
{
"operation": "Highlighting deleted",
"time": "37.597",
"value": "Garra regressus_cell_0_7_1"
},
{
"operation": "Highlighting deleted",
"time": "37.977",
"value": "Garra regressus_cell_0_8_1"
},
{
"operation": "Highlighting deleted",
"time": "38.67",
"value": "Garra regressus_cell_0_7_0"
},
{
"operation": "Highlighting deleted",
"time": "39.08",
"value": "Garra regressus_cell_0_6_0"
},
{
"operation": "Highlighting deleted",
"time": "39.082",
"value": "Garra regressus_title"
},
{
"operation": "Highlighting deleted",
"time": "39.085",
"value": "Garra regressus_header_cell_0_2_0"
},
{
"operation": "Highlighting deleted",
"time": "39.087",
"value": "Garra regressus_header_cell_0_1_0"
},
{
"operation": "Highlighting deleted",
"time": "39.088",
"value": "Garra regressus_header_cell_0_0_0"
},
{
"operation": "finish",
"time": "57.675",
"value": "finish"
}
]
}
|
31122
|
1937–38 AHL season saw Jack Markle top the scoring list for the season with 54 points, followed by Eddie Convey with 52, and Lorne Duguid in third with 49 points.
|
{
"evidence": [
"Jack Markle_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:\n1937–38 AHL season saw Jack Markle top the scoring list for the season with 54 points, followed by Eddie Convey with 52, and Lorne Duguid in third with 49 points.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Jack Markle_sentence_1"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7606,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.666",
"value": "?search="
},
{
"operation": "search",
"time": "24.227",
"value": "Jack Markle"
},
{
"operation": "Now on",
"time": "25.083",
"value": "Jack Markle"
},
{
"operation": "Highlighting",
"time": "47.669",
"value": "Jack Markle_sentence_1"
},
{
"operation": "search",
"time": "57.083",
"value": "Toronto Maple Leafs"
},
{
"operation": "Now on",
"time": "58.187",
"value": "Toronto Maple Leafs"
},
{
"operation": "search",
"time": "117.194",
"value": "AHL"
},
{
"operation": "Now on",
"time": "118.181",
"value": "AHL"
},
{
"operation": "hyperlink",
"time": "127.52",
"value": "American Hockey League"
},
{
"operation": "Now on",
"time": "128.518",
"value": "American Hockey League"
},
{
"operation": "search",
"time": "214.813",
"value": "Jack Markle"
},
{
"operation": "Now on",
"time": "215.409",
"value": "Jack Markle"
},
{
"operation": "hyperlink",
"time": "240.453",
"value": "National Hockey League"
},
{
"operation": "Now on",
"time": "241.831",
"value": "National Hockey League"
},
{
"operation": "search",
"time": "331.802",
"value": "American Hockey League"
},
{
"operation": "Now on",
"time": "332.564",
"value": "American Hockey League"
},
{
"operation": "finish",
"time": "446.135",
"value": "finish"
}
]
}
|
92507
|
Keno is a turnover tax that is apart of the major forms of gambling taxation in Australia.
|
{
"evidence": [
"Keno_sentence_1",
"Lottery_sentence_14",
"Lottery_sentence_80",
"Lottery_sentence_81",
"Lottery_sentence_82",
"Lottery_sentence_83",
"Lottery_sentence_84"
],
"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:\nKeno is a turnover tax that is apart of the major forms of gambling taxation in Australia.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Keno_sentence_1",
"Lottery_sentence_14",
"Lottery_sentence_80",
"Lottery_sentence_81",
"Lottery_sentence_82",
"Lottery_sentence_83",
"Lottery_sentence_84"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7607,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.618",
"value": "?search="
},
{
"operation": "search",
"time": "10.627",
"value": "Keno"
},
{
"operation": "Now on",
"time": "11.693",
"value": "Keno"
},
{
"operation": "Highlighting",
"time": "18.214",
"value": "Keno_sentence_1"
},
{
"operation": "search",
"time": "33.169",
"value": "Contains...gambling taxation in Australia"
},
{
"operation": "Now on",
"time": "35.127",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "42.256",
"value": "Lottery"
},
{
"operation": "Now on",
"time": "43.555",
"value": "Lottery"
},
{
"operation": "hyperlink",
"time": "56.081",
"value": "5.1 Australia Lottery_section_10"
},
{
"operation": "hyperlink",
"time": "67.359",
"value": "Keno"
},
{
"operation": "Now on",
"time": "68.16",
"value": "Keno"
},
{
"operation": "back-button-clicked",
"time": "80.448",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "81.02",
"value": "Lottery#Australia Lottery section 10"
},
{
"operation": "Highlighting",
"time": "119.502",
"value": "Lottery_sentence_14"
},
{
"operation": "hyperlink",
"time": "127.063",
"value": "5.1 Australia Lottery_section_10"
},
{
"operation": "Highlighting",
"time": "128.733",
"value": "Lottery_sentence_80"
},
{
"operation": "Highlighting",
"time": "129.229",
"value": "Lottery_sentence_81"
},
{
"operation": "Highlighting",
"time": "130.467",
"value": "Lottery_sentence_82"
},
{
"operation": "Highlighting",
"time": "130.977",
"value": "Lottery_sentence_83"
},
{
"operation": "Highlighting",
"time": "131.552",
"value": "Lottery_sentence_84"
},
{
"operation": "finish",
"time": "138.822",
"value": "finish"
}
]
}
|
exfever_5163
|
Sir Ben Kingsley is an English actor who has garnered numerous accolades, including an award presented by the Recording Academy to recognize the "Outstanding Achievement in the music industry" of the United States.
|
{
"evidence": [
"Ben Kingsley",
"Grammy Award"
],
"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 Ben Kingsley is an English actor who has garnered numerous accolades, including an award presented by the Recording Academy to recognize the \"Outstanding Achievement in the music industry\" of the United States.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Ben Kingsley",
"Grammy Award"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7608,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
78627
|
Spyridon Valettas from 1827 to 1830 stayed in Rijeka.
|
{
"evidence": [
"Spyridon Valettas_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:\nSpyridon Valettas from 1827 to 1830 stayed in Rijeka.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Spyridon Valettas_sentence_7"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7609,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.901",
"value": "?search="
},
{
"operation": "search",
"time": "9.425",
"value": "Spyridon Valettas"
},
{
"operation": "Now on",
"time": "10.522",
"value": "Spyridon Valettas"
},
{
"operation": "Page search",
"time": "23.822",
"value": "Rijeka"
},
{
"operation": "Highlighting",
"time": "27.789",
"value": "Spyridon Valettas_sentence_7"
},
{
"operation": "finish",
"time": "46.884",
"value": "finish"
}
]
}
|
exfever_7844
|
John Constantine was the first antihero who appears in American comic books.
|
{
"evidence": [
"John Constantine",
"Alan Moore"
],
"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 Constantine was the first antihero who appears in American comic books.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"John Constantine",
"Alan Moore"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7610,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6424
|
Jane Eyre is a 2011 romantic drama film written by Moira Buffini based on a novel by Canadian writer Charlotte Brontë, published under the pen name "Currer Bell" on 16 October 1847 by Smith, Elder & Co. of London.
|
{
"evidence": [
"Jane Eyre (2011 film)",
"Jane Eyre"
],
"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:\nJane Eyre is a 2011 romantic drama film written by Moira Buffini based on a novel by Canadian writer Charlotte Brontë, published under the pen name \"Currer Bell\" on 16 October 1847 by Smith, Elder & Co. of London.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Jane Eyre (2011 film)",
"Jane Eyre"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7611,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4236
|
Roswell is an American science fiction television series that was moved to an American broadcast television network with another American television network as its competitor.
|
{
"evidence": [
"Roswell (TV series)",
"UPN",
"The WB"
],
"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:\nRoswell is an American science fiction television series that was moved to an American broadcast television network with another American television network as its competitor.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Roswell (TV series)",
"UPN",
"The WB"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7612,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_998
|
The Midwestern United States is between the Western and Northern U.S. where a state in the U.S. bordered by one of the five Great Lakes of North America is located.
|
{
"evidence": [
"Midwestern United States",
"Indiana",
"Lake Michigan"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Midwestern United States is between the Western and Northern U.S. where a state in the U.S. bordered by one of the five Great Lakes of North America is located.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Midwestern United States",
"Indiana",
"Lake Michigan"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7613,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
14988
|
For exiled Vietnamese who have been displaced in overseas communities, Reunification Day is just another day.
|
{
"evidence": [
"Reunification Day_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:\nFor exiled Vietnamese who have been displaced in overseas communities, Reunification Day is just another day.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Reunification Day_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7614,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.885",
"value": "?search="
},
{
"operation": "search",
"time": "18.208",
"value": "Reunification Day"
},
{
"operation": "Now on",
"time": "18.875",
"value": "Reunification Day"
},
{
"operation": "Highlighting",
"time": "47.45",
"value": "Reunification Day_sentence_4"
},
{
"operation": "finish",
"time": "79.263",
"value": "finish"
}
]
}
|
exfever_4709
|
50 Cent is a Jamaican rapper known as the "master of the nuanced art of lyrical brevity" was born in the South Jamaica neighborhood of the largest borough of New York City located on Long Island.
|
{
"evidence": [
"50 Cent",
"Queens"
],
"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:\n50 Cent is a Jamaican rapper known as the \"master of the nuanced art of lyrical brevity\" was born in the South Jamaica neighborhood of the largest borough of New York City located on Long Island.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"50 Cent",
"Queens"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7615,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
52241
|
Áedán mac Gabráin was a son of Gabrán mac Domangairt and ruled the kingdom of Dál Riata situated in modern England.
|
{
"evidence": [
"Áedán mac Gabráin_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:\nÁedán mac Gabráin was a son of Gabrán mac Domangairt and ruled the kingdom of Dál Riata situated in modern England.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Áedán mac Gabráin_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7616,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.119",
"value": "?search="
},
{
"operation": "search",
"time": "10.729",
"value": "Áedán mac Gabráin"
},
{
"operation": "Now on",
"time": "11.584",
"value": "Áedán mac Gabráin"
},
{
"operation": "Highlighting",
"time": "27.04",
"value": "Áedán mac Gabráin_sentence_2"
},
{
"operation": "Highlighting",
"time": "35.924",
"value": "Áedán mac Gabráin_sentence_0"
},
{
"operation": "Highlighting",
"time": "46.476",
"value": "Áedán mac Gabráin_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "49.07",
"value": "Áedán mac Gabráin_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "50.322",
"value": "Áedán mac Gabráin_sentence_0"
},
{
"operation": "finish",
"time": "57.432",
"value": "finish"
}
]
}
|
exfever_8921
|
Audrey Horne is the second character created by David Lynch.
|
{
"evidence": [
"Audrey Horne",
"American Broadcasting Company"
],
"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:\nAudrey Horne is the second character created by David Lynch.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Audrey Horne",
"American Broadcasting Company"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7617,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
55800
|
In the 2018 season the Chicago Sky won all their games until July when the winning streak ended.
|
{
"evidence": [
"2018 Chicago Sky season_sentence_2",
"2018 Chicago Sky season_sentence_3",
"2018 Chicago Sky season_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:\nIn the 2018 season the Chicago Sky won all their games until July when the winning streak ended.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"2018 Chicago Sky season_sentence_2",
"2018 Chicago Sky season_sentence_3",
"2018 Chicago Sky season_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7618,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.827",
"value": "?search="
},
{
"operation": "search",
"time": "10.885",
"value": "2018 Chicago Sky season"
},
{
"operation": "Now on",
"time": "12.462",
"value": "2018 Chicago Sky season"
},
{
"operation": "Highlighting",
"time": "24.888",
"value": "2018 Chicago Sky season_sentence_3"
},
{
"operation": "Highlighting",
"time": "26.399",
"value": "2018 Chicago Sky season_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "30.251",
"value": "2018 Chicago Sky season_sentence_3"
},
{
"operation": "Highlighting",
"time": "38.435",
"value": "2018 Chicago Sky season_sentence_3"
},
{
"operation": "hyperlink",
"time": "42.609",
"value": "Indiana"
},
{
"operation": "Now on",
"time": "43.698",
"value": "2018 Indiana Fever season"
},
{
"operation": "back-button-clicked",
"time": "44.203",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "44.346",
"value": "2018 Chicago Sky season"
},
{
"operation": "Highlighting",
"time": "46.188",
"value": "2018 Chicago Sky season_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "61.106",
"value": "2018 Chicago Sky season_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "61.765",
"value": "2018 Chicago Sky season_sentence_3"
},
{
"operation": "Highlighting",
"time": "62.448",
"value": "2018 Chicago Sky season_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "65.606",
"value": "2018 Chicago Sky season_sentence_4"
},
{
"operation": "Highlighting",
"time": "66.408",
"value": "2018 Chicago Sky season_sentence_3"
},
{
"operation": "Highlighting",
"time": "68.083",
"value": "2018 Chicago Sky season_sentence_2"
},
{
"operation": "Highlighting",
"time": "71.956",
"value": "2018 Chicago Sky season_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "75.281",
"value": "2018 Chicago Sky season_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "75.72",
"value": "2018 Chicago Sky season_sentence_3"
},
{
"operation": "Highlighting",
"time": "76.152",
"value": "2018 Chicago Sky season_sentence_4"
},
{
"operation": "finish",
"time": "79.983",
"value": "finish"
}
]
}
|
54930
|
Premiership Rugby top scorers had ten names between 1997 and 2020, one of which is Ollie Thorley(born 11 September 1996).
|
{
"evidence": [
"Premiership Rugby top scorers_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:\nPremiership Rugby top scorers had ten names between 1997 and 2020, one of which is Ollie Thorley(born 11 September 1996).\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Premiership Rugby top scorers_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7619,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.128",
"value": "?search="
},
{
"operation": "search",
"time": "10.128",
"value": "Premiership Rugby top scorers"
},
{
"operation": "Now on",
"time": "11.561",
"value": "Premiership Rugby top scorers"
},
{
"operation": "Highlighting",
"time": "56.887",
"value": "Premiership Rugby top scorers_cell_2_1_0"
},
{
"operation": "Highlighting",
"time": "56.888",
"value": "Premiership Rugby top scorers_section_0"
},
{
"operation": "Highlighting",
"time": "56.89",
"value": "Premiership Rugby top scorers_header_cell_2_0_0"
},
{
"operation": "Highlighting deleted",
"time": "65.151",
"value": "Premiership Rugby top scorers_cell_2_1_0"
},
{
"operation": "Highlighting deleted",
"time": "65.154",
"value": "Premiership Rugby top scorers_section_0"
},
{
"operation": "Highlighting deleted",
"time": "65.155",
"value": "Premiership Rugby top scorers_header_cell_2_0_0"
},
{
"operation": "Highlighting",
"time": "166.115",
"value": "Premiership Rugby top scorers_sentence_2"
},
{
"operation": "finish",
"time": "173.66",
"value": "finish"
}
]
}
|
exfever_1674
|
The Boston Latin School is a public exam school in Boston, in the most populous state in the New England region of the United States.
|
{
"evidence": [
"Boston Latin School",
"Massachusetts"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Boston Latin School is a public exam school in Boston, in the most populous state in the New England region of the United States.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Boston Latin School",
"Massachusetts"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7620,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6616
|
Lymelife is a 2008 comedy-drama film directed by Derick Martini and executive produced by an American film director and screenwriter who has received various honors in 2017, including the award established on February 26, 1974, by the board of directors of the American Film Institute.
|
{
"evidence": [
"Lymelife",
"Martin Scorsese",
"AFI Life Achievement Award"
],
"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:\nLymelife is a 2008 comedy-drama film directed by Derick Martini and executive produced by an American film director and screenwriter who has received various honors in 2017, including the award established on February 26, 1974, by the board of directors of the American Film Institute.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Lymelife",
"Martin Scorsese",
"AFI Life Achievement Award"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7621,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
51395
|
ADH1B encodes a protein in the alcohol dehydrogenase family, a group of enzymes that occur in many organisms.
|
{
"evidence": [
"ADH1B_sentence_1",
"Alcohol dehydrogenase_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:\nADH1B encodes a protein in the alcohol dehydrogenase family, a group of enzymes that occur in many organisms.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"ADH1B_sentence_1",
"Alcohol dehydrogenase_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7622,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.285",
"value": "?search="
},
{
"operation": "search",
"time": "93.213",
"value": "ADH1B"
},
{
"operation": "Now on",
"time": "94.17",
"value": "ADH1B"
},
{
"operation": "Highlighting",
"time": "676.945",
"value": "ADH1B_sentence_1"
},
{
"operation": "search",
"time": "910.719",
"value": "Alcohol dehydrogenase"
},
{
"operation": "Now on",
"time": "913.067",
"value": "Alcohol dehydrogenase"
},
{
"operation": "Highlighting",
"time": "928.533",
"value": "Alcohol dehydrogenase_sentence_0"
},
{
"operation": "finish",
"time": "954.878",
"value": "finish"
}
]
}
|
exfever_7362
|
Syco is an American media and entertainment company, formed in 2005 through a company founded in 1929 as the American Record Corporation.
|
{
"evidence": [
"Syco",
"Sony Music"
],
"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:\nSyco is an American media and entertainment company, formed in 2005 through a company founded in 1929 as the American Record Corporation.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Syco",
"Sony Music"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7623,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1537
|
Fences is a 2016 American period drama film starring, produced and directed by Denzel Washington and written by August Wilson, based on his 1985 play of the same name which earned award for achievements in newspaper, magazine, online journalism, literature and musical composition within the United States.
|
{
"evidence": [
"Fences (film)",
"Pulitzer Prize"
],
"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:\nFences is a 2016 American period drama film starring, produced and directed by Denzel Washington and written by August Wilson, based on his 1985 play of the same name which earned award for achievements in newspaper, magazine, online journalism, literature and musical composition within the United States.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Fences (film)",
"Pulitzer Prize"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7624,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
5660
|
Wetlet Township is a township of the Shwebo district of the Sagaing region of Burma, a country which is also known as Myanmar.
|
{
"evidence": [
"Wetlet Township_sentence_0",
"Myanmar_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:\nWetlet Township is a township of the Shwebo district of the Sagaing region of Burma, a country which is also known as Myanmar.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Wetlet Township_sentence_0",
"Myanmar_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7625,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.98",
"value": "?search="
},
{
"operation": "search",
"time": "13.257",
"value": "Wetlet Township"
},
{
"operation": "Now on",
"time": "14.3",
"value": "Wetlet Township"
},
{
"operation": "Highlighting",
"time": "19.318",
"value": "Wetlet Township_sentence_0"
},
{
"operation": "hyperlink",
"time": "32.724",
"value": "Burma (Myanmar)"
},
{
"operation": "Now on",
"time": "34.76",
"value": "Myanmar"
},
{
"operation": "Highlighting",
"time": "37.476",
"value": "Myanmar_sentence_5"
},
{
"operation": "finish",
"time": "44.487",
"value": "finish"
}
]
}
|
exfever_2953
|
The Apple Store is a chain of retail stores that sells a digital media player and microconsole that is developed and marketed by Apple Inc.
|
{
"evidence": [
"Apple Store",
"Apple TV"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Apple Store is a chain of retail stores that sells a digital media player and microconsole that is developed and marketed by Apple Inc.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Apple Store",
"Apple TV"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7626,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3551
|
The Messenger is a 2009 war drama film starring an American actor who first became known for his role as bartender Woody Boyd on the NBC sitcom Cheers.
|
{
"evidence": [
"The Messenger (2009 film)",
"Woody Harrelson"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Messenger is a 2009 war drama film starring an American actor who first became known for his role as bartender Woody Boyd on the NBC sitcom Cheers.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Messenger (2009 film)",
"Woody Harrelson"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7627,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
43258
|
Yuichi Nakamaru was born in Kita-ku but raised in Sumida-ku, Tokyo as the oldest child in a family of five that included his parents and two younger brothers.
|
{
"evidence": [
"Yuichi Nakamaru_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:\nYuichi Nakamaru was born in Kita-ku but raised in Sumida-ku, Tokyo as the oldest child in a family of five that included his parents and two younger brothers.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Yuichi Nakamaru_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7628,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.706",
"value": "?search="
},
{
"operation": "search",
"time": "12.524",
"value": "Yuichi Nakamaru"
},
{
"operation": "Now on",
"time": "13.497",
"value": "Yuichi Nakamaru"
},
{
"operation": "Highlighting",
"time": "33.92",
"value": "Yuichi Nakamaru_sentence_0"
},
{
"operation": "Page search",
"time": "43.01",
"value": "sumida"
},
{
"operation": "Highlighting",
"time": "49.48",
"value": "Yuichi Nakamaru_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "52.83",
"value": "Yuichi Nakamaru_sentence_0"
},
{
"operation": "Page search",
"time": "65.699",
"value": "oldest"
},
{
"operation": "Highlighting",
"time": "71.449",
"value": "Yuichi Nakamaru_sentence_4"
},
{
"operation": "hyperlink",
"time": "100.523",
"value": "Sumida-ku, Tokyo"
},
{
"operation": "Now on",
"time": "101.381",
"value": "Sumida-ku, Tokyo"
},
{
"operation": "back-button-clicked",
"time": "104.242",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "104.506",
"value": "Yuichi Nakamaru"
},
{
"operation": "Highlighting deleted",
"time": "107.051",
"value": "Yuichi Nakamaru_sentence_3"
},
{
"operation": "finish",
"time": "223.163",
"value": "finish"
}
]
}
|
exfever_3019
|
Comedy films was popularized by Jerry Lewis.
|
{
"evidence": [
"Jerry Lewis",
"Popular culture"
],
"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:\nComedy films was popularized by Jerry Lewis.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Jerry Lewis",
"Popular culture"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7629,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6968
|
The Vampire Diaries is an American drama television series developed by a Canadian screenwriter, who is also known for creating the teen drama series Dawson's Creek for an American television network, first launched on broadcast television on January 11, 1995.
|
{
"evidence": [
"The Vampire Diaries",
"Kevin Williamson (screenwriter)",
"The WB"
],
"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 Vampire Diaries is an American drama television series developed by a Canadian screenwriter, who is also known for creating the teen drama series Dawson's Creek for an American television network, first launched on broadcast television on January 11, 1995.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Vampire Diaries",
"Kevin Williamson (screenwriter)",
"The WB"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7630,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6082
|
Wayne Rooney is a former football player who has the most goals scored for the?national football team that has represented Japan.
|
{
"evidence": [
"Wayne Rooney",
"England national football team"
],
"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:\nWayne Rooney is a former football player who has the most goals scored for the?national football team that has represented Japan.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Wayne Rooney",
"England national football team"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7631,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
79263
|
Norm Fieldgate Trophy (The trophy named after former BC Lions linebacker Norm Fieldgate) is a Canadian Football League trophy awarded to the outstanding defensive player in the West Division.
|
{
"evidence": [
"Norm Fieldgate Trophy_sentence_3",
"Norm Fieldgate Trophy_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:\nNorm Fieldgate Trophy (The trophy named after former BC Lions linebacker Norm Fieldgate) is a Canadian Football League trophy awarded to the outstanding defensive player in the West Division.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Norm Fieldgate Trophy_sentence_3",
"Norm Fieldgate Trophy_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7632,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.729",
"value": "?search="
},
{
"operation": "search",
"time": "13.632",
"value": "Norm Fieldgate Trophy"
},
{
"operation": "Now on",
"time": "14.526",
"value": "Norm Fieldgate Trophy"
},
{
"operation": "Highlighting",
"time": "41.852",
"value": "Norm Fieldgate Trophy_sentence_3"
},
{
"operation": "Highlighting",
"time": "47.119",
"value": "Norm Fieldgate Trophy_sentence_0"
},
{
"operation": "finish",
"time": "70.874",
"value": "finish"
}
]
}
|
71045
|
Carroll County Broadcasting Company retained ownership of WTTR AM and FM, until the station was sold to Shamrock Broadcasting in the early 1980s; then Dwight Dingle, known for decades as the radio "Voice of Baltimore County", managed WTTR from 1974 until his death.
|
{
"evidence": [
"WTTR_sentence_14",
"WTTR_sentence_15"
],
"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:\nCarroll County Broadcasting Company retained ownership of WTTR AM and FM, until the station was sold to Shamrock Broadcasting in the early 1980s; then Dwight Dingle, known for decades as the radio \"Voice of Baltimore County\", managed WTTR from 1974 until his death.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"WTTR_sentence_14",
"WTTR_sentence_15"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7633,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.962",
"value": "?search="
},
{
"operation": "search",
"time": "17.874",
"value": "WTTR AM and FM,"
},
{
"operation": "Now on",
"time": "19.031",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "21.802",
"value": "WTTR"
},
{
"operation": "Now on",
"time": "22.364",
"value": "WTTR"
},
{
"operation": "Highlighting",
"time": "59.471",
"value": "WTTR_sentence_14"
},
{
"operation": "Highlighting",
"time": "72.607",
"value": "WTTR_sentence_15"
},
{
"operation": "search",
"time": "94.472",
"value": "Dwight Dingle"
},
{
"operation": "Now on",
"time": "97.513",
"value": "undefined"
},
{
"operation": "search",
"time": "124.766",
"value": "Voice of Baltimore"
},
{
"operation": "Now on",
"time": "129.105",
"value": "undefined"
},
{
"operation": "search",
"time": "156.028",
"value": "Baltimore COunty"
},
{
"operation": "Now on",
"time": "157.764",
"value": "Baltimore County, Maryland"
},
{
"operation": "Page search",
"time": "165.769",
"value": "Dwight"
},
{
"operation": "search",
"time": "222.053",
"value": "Dwight Dingle"
},
{
"operation": "Now on",
"time": "223.079",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "225.646",
"value": "WTTR"
},
{
"operation": "Now on",
"time": "226.081",
"value": "WTTR"
},
{
"operation": "finish",
"time": "265.078",
"value": "finish"
}
]
}
|
64815
|
Friendship Heights station is a Washington Metro station was formerly known as the Metrorail and is a rapid transit system serving the Washington metropolitan area of the United States.
|
{
"evidence": [
"Friendship Heights station_sentence_0",
"Washington Metro_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:\nFriendship Heights station is a Washington Metro station was formerly known as the Metrorail and is a rapid transit system serving the Washington metropolitan area of the United States.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Friendship Heights station_sentence_0",
"Washington Metro_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7634,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.079",
"value": "?search="
},
{
"operation": "search",
"time": "13.532",
"value": "Friendship Heights station"
},
{
"operation": "Now on",
"time": "25.656",
"value": "Friendship Heights station"
},
{
"operation": "Highlighting",
"time": "29.785",
"value": "Friendship Heights station_sentence_0"
},
{
"operation": "Page search",
"time": "51.887",
"value": "Metrorail"
},
{
"operation": "hyperlink",
"time": "79.346",
"value": "Washington Metro"
},
{
"operation": "Now on",
"time": "81.651",
"value": "Washington Metro"
},
{
"operation": "Highlighting",
"time": "85.082",
"value": "Washington Metro_sentence_1"
},
{
"operation": "finish",
"time": "107.274",
"value": "finish"
}
]
}
|
exfever_6138
|
Joffrey Baratheon is a fictional character in the A Song of Ice and Fire series of epic fantasy novels which is authored by GRRM and was adapted into an American fantasy drama television series created by David Benioff and D. B. Weiss for HBO.
|
{
"evidence": [
"Joffrey Baratheon",
"George R. R. Martin",
"Game of Thrones"
],
"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:\nJoffrey Baratheon is a fictional character in the A Song of Ice and Fire series of epic fantasy novels which is authored by GRRM and was adapted into an American fantasy drama television series created by David Benioff and D. B. Weiss for HBO.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Joffrey Baratheon",
"George R. R. Martin",
"Game of Thrones"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7635,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
25268
|
The team with the most points will play the second leg home, while the bottom two teams on overall classification will be relegated.
|
{
"evidence": [
"2012 Campeonato Paranaense_sentence_9",
"2012 Campeonato Paranaense_sentence_10"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe team with the most points will play the second leg home, while the bottom two teams on overall classification will be relegated.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"2012 Campeonato Paranaense_sentence_9",
"2012 Campeonato Paranaense_sentence_10"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7636,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.218",
"value": "?search="
},
{
"operation": "search",
"time": "18.963",
"value": "Contains...The team with the most points will play the second leg home, while the bottom two teams on overall classification will be relegated"
},
{
"operation": "Now on",
"time": "22.976",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "37.923",
"value": "2013 Campeonato Paranaense"
},
{
"operation": "Now on",
"time": "39.093",
"value": "2013 Campeonato Paranaense"
},
{
"operation": "hyperlink",
"time": "42.054",
"value": "Campeonato Paranaense de Futebol Profissional da 1ª Divisão"
},
{
"operation": "Now on",
"time": "43.411",
"value": "Campeonato Paranaense"
},
{
"operation": "Page search",
"time": "63.04",
"value": "while the bottom two teams on overall classification will be relegated"
},
{
"operation": "back-button-clicked",
"time": "71.626",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "72.067",
"value": "2013 Campeonato Paranaense"
},
{
"operation": "Page search",
"time": "93.794",
"value": "while the bottom two teams on overall classification will be relegated"
},
{
"operation": "Page search",
"time": "111.052",
"value": "while the bottom two teams on overall classification will be relegated"
},
{
"operation": "back-button-clicked",
"time": "120.475",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "120.509",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "122.188",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "125.277",
"value": "Participating teams 2013 Campeonato Paranaense_section_1"
},
{
"operation": "Now on",
"time": "126.119",
"value": "2013 Campeonato Paranaense#Participating teams 2013 Campeonato Paranaense section 1"
},
{
"operation": "back-button-clicked",
"time": "129.871",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "131.195",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "135.548",
"value": "Overall classification 2012 Campeonato Paranaense_section_6"
},
{
"operation": "Now on",
"time": "137.075",
"value": "2012 Campeonato Paranaense#Overall classification 2012 Campeonato Paranaense section 6"
},
{
"operation": "Highlighting",
"time": "178.429",
"value": "2012 Campeonato Paranaense_sentence_5"
},
{
"operation": "Highlighting",
"time": "187.625",
"value": "2012 Campeonato Paranaense_sentence_9"
},
{
"operation": "Highlighting",
"time": "188.787",
"value": "2012 Campeonato Paranaense_sentence_10"
},
{
"operation": "Highlighting deleted",
"time": "189.867",
"value": "2012 Campeonato Paranaense_sentence_5"
},
{
"operation": "finish",
"time": "205.748",
"value": "finish"
}
]
}
|
exfever_1959
|
Tardigrades are a phylum of six-legged segmented micro-animals that were named by the Italian biologist who did research on biogenesis that paved the way for the downfall of the theory of spontaneous generation.
|
{
"evidence": [
"Tardigrade",
"Lazzaro Spallanzani"
],
"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:\nTardigrades are a phylum of six-legged segmented micro-animals that were named by the Italian biologist who did research on biogenesis that paved the way for the downfall of the theory of spontaneous generation.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Tardigrade",
"Lazzaro Spallanzani"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7637,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
51977
|
By the late 19th century, the lock dam between Chamberlain and Eagle Lakes (the first, largest, and deepest lake of the Allagash Wilderness Waterway in the North Maine Woods) had become a hindrance to logging operations; so to speed up the movement of timbers between the two lakes, the cable-drawn steam-powered tramway was implemented.
|
{
"evidence": [
"Eagle Lake Tramway_sentence_7",
"Eagle Lake Tramway_sentence_8",
"Eagle Lake (Maine)_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:\nBy the late 19th century, the lock dam between Chamberlain and Eagle Lakes (the first, largest, and deepest lake of the Allagash Wilderness Waterway in the North Maine Woods) had become a hindrance to logging operations; so to speed up the movement of timbers between the two lakes, the cable-drawn steam-powered tramway was implemented.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Eagle Lake Tramway_sentence_7",
"Eagle Lake Tramway_sentence_8",
"Eagle Lake (Maine)_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7638,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.96",
"value": "?search="
},
{
"operation": "search",
"time": "19.005",
"value": "Chamberlain and Eagle Lakes"
},
{
"operation": "Now on",
"time": "20.875",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "28.936",
"value": "Description and history Eagle Lake Tramway_section_0"
},
{
"operation": "Now on",
"time": "29.506",
"value": "Eagle Lake Tramway#Description and history Eagle Lake Tramway section 0"
},
{
"operation": "Highlighting",
"time": "39.441",
"value": "Eagle Lake Tramway_sentence_7"
},
{
"operation": "Highlighting",
"time": "45.156",
"value": "Eagle Lake Tramway_sentence_8"
},
{
"operation": "Highlighting deleted",
"time": "63.47",
"value": "Eagle Lake Tramway_sentence_7"
},
{
"operation": "Highlighting",
"time": "65.28",
"value": "Eagle Lake Tramway_sentence_7"
},
{
"operation": "Highlighting deleted",
"time": "66.225",
"value": "Eagle Lake Tramway_sentence_8"
},
{
"operation": "Highlighting",
"time": "66.587",
"value": "Eagle Lake Tramway_sentence_8"
},
{
"operation": "search",
"time": "97.965",
"value": "Allagash Wilderness Waterway"
},
{
"operation": "Now on",
"time": "98.887",
"value": "Allagash Wilderness Waterway"
},
{
"operation": "search",
"time": "112.82",
"value": "Contains...Eagle Lakes"
},
{
"operation": "Now on",
"time": "114.273",
"value": "undefined"
},
{
"operation": "search",
"time": "157.92",
"value": "Eagle Lake, Maine"
},
{
"operation": "Now on",
"time": "159.053",
"value": "Eagle Lake, Maine"
},
{
"operation": "search",
"time": "183.538",
"value": "Eagle Lake (Maine)"
},
{
"operation": "Now on",
"time": "184.665",
"value": "Eagle Lake (Maine)"
},
{
"operation": "Highlighting",
"time": "186.75",
"value": "Eagle Lake (Maine)_sentence_1"
},
{
"operation": "finish",
"time": "192.135",
"value": "finish"
}
]
}
|
exfever_2150
|
Nick Cannon is an American actor who acted in the films Drumline, Love Don't Cost a Thing, and a 2005 American comedy-drama directed by Malcolm D. Lee.
|
{
"evidence": [
"Nick Cannon",
"Roll Bounce"
],
"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:\nNick Cannon is an American actor who acted in the films Drumline, Love Don't Cost a Thing, and a 2005 American comedy-drama directed by Malcolm D. Lee.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Nick Cannon",
"Roll Bounce"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7639,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5759
|
Lyndia Hearst is a great-granddaughter of newspaper publisher and politician who also entered the publishing business in 1878 with Mitchell Trubitt.
|
{
"evidence": [
"Lydia Hearst",
"William Randolph Hearst"
],
"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:\nLyndia Hearst is a great-granddaughter of newspaper publisher and politician who also entered the publishing business in 1878 with Mitchell Trubitt.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Lydia Hearst",
"William Randolph Hearst"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7640,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
36901
|
In the UEFA Euro 1988 group match, the Irish and Soviets led the group after two games through a 1–1 draw in Hanover wherein Ronnie Whelan scored a spectacular left-foot volley from 18 yards to put the Irish into the lead.
|
{
"evidence": [
"UEFA Euro 1988_sentence_44",
"UEFA Euro 1988_sentence_45"
],
"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:\nIn the UEFA Euro 1988 group match, the Irish and Soviets led the group after two games through a 1–1 draw in Hanover wherein Ronnie Whelan scored a spectacular left-foot volley from 18 yards to put the Irish into the lead.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"UEFA Euro 1988_sentence_44",
"UEFA Euro 1988_sentence_45"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7641,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.701",
"value": "?search="
},
{
"operation": "search",
"time": "18.296",
"value": "UEFA Euro 1988"
},
{
"operation": "Now on",
"time": "19.525",
"value": "UEFA Euro 1988"
},
{
"operation": "hyperlink",
"time": "32.288",
"value": "7 Group stage UEFA Euro 1988_section_9"
},
{
"operation": "hyperlink",
"time": "57.431",
"value": "UEFA Euro 1988 Group 1"
},
{
"operation": "Now on",
"time": "58.028",
"value": "UEFA Euro 1988 Group 1"
},
{
"operation": "back-button-clicked",
"time": "67.661",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "67.944",
"value": "UEFA Euro 1988#Group stage UEFA Euro 1988 section 9"
},
{
"operation": "search",
"time": "91.694",
"value": "Contains...UEFA Euro 1988 group match"
},
{
"operation": "Now on",
"time": "93.523",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "104.298",
"value": "Group matches UEFA Euro 1988_section_2"
},
{
"operation": "Now on",
"time": "104.962",
"value": "UEFA Euro 1988#Group matches UEFA Euro 1988 section 2"
},
{
"operation": "Highlighting",
"time": "177.611",
"value": "UEFA Euro 1988_sentence_44"
},
{
"operation": "Highlighting",
"time": "184.198",
"value": "UEFA Euro 1988_sentence_45"
},
{
"operation": "finish",
"time": "194.19",
"value": "finish"
}
]
}
|
exfever_5770
|
Joaquin Phoenix won an award presented by the Recording Academy.
|
{
"evidence": [
"Joaquin Phoenix",
"Grammy Award"
],
"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:\nJoaquin Phoenix won an award presented by the Recording Academy.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Joaquin Phoenix",
"Grammy Award"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7642,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2956
|
The first inauguration of Bill Clinton as the 44th president of the United States was held at the West Front of the meeting place of the United States Congress and the seat of the legislative branch of the U.S. federal government.
|
{
"evidence": [
"First inauguration of Bill Clinton",
"United States Capitol"
],
"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 first inauguration of Bill Clinton as the 44th president of the United States was held at the West Front of the meeting place of the United States Congress and the seat of the legislative branch of the U.S. federal government.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"First inauguration of Bill Clinton",
"United States Capitol"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7643,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
70484
|
Vendor is a supply chain management term where they can only sell B2B.
|
{
"evidence": [
"Vendor_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:\nVendor is a supply chain management term where they can only sell B2B.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Vendor_sentence_5"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7644,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.989",
"value": "?search="
},
{
"operation": "search",
"time": "15.693",
"value": "Vendor"
},
{
"operation": "Now on",
"time": "16.797",
"value": "Vendor"
},
{
"operation": "Highlighting",
"time": "27.06",
"value": "Vendor_sentence_4"
},
{
"operation": "Highlighting",
"time": "33.86",
"value": "Vendor_sentence_5"
},
{
"operation": "Highlighting deleted",
"time": "35.98",
"value": "Vendor_sentence_4"
},
{
"operation": "finish",
"time": "40.939",
"value": "finish"
}
]
}
|
exfever_8362
|
A cappella is a performance of music used in the film Glee.
|
{
"evidence": [
"Buddy Holly",
"Gospel music",
"A cappella"
],
"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:\nA cappella is a performance of music used in the film Glee.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Buddy Holly",
"Gospel music",
"A cappella"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7645,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
85568
|
Sam68 not only regulates the alternatve splicing of a number of cancer-related genes, but also interacts with splicing complex proteins KHDRBS3 (T-STAR) and Metadherin (involved in HIF-1alpha mediated angiogenesis), which also alter CD44 splicing.
|
{
"evidence": [
"KHDRBS1_sentence_12",
"KHDRBS1_sentence_15",
"MTDH_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:\nSam68 not only regulates the alternatve splicing of a number of cancer-related genes, but also interacts with splicing complex proteins KHDRBS3 (T-STAR) and Metadherin (involved in HIF-1alpha mediated angiogenesis), which also alter CD44 splicing.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"KHDRBS1_sentence_12",
"KHDRBS1_sentence_15",
"MTDH_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7646,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.211",
"value": "?search="
},
{
"operation": "search",
"time": "12.659",
"value": "Contains...Sam68"
},
{
"operation": "Now on",
"time": "13.804",
"value": "undefined"
},
{
"operation": "search",
"time": "48.051",
"value": "KHDRBS1"
},
{
"operation": "Now on",
"time": "49.388",
"value": "KHDRBS1"
},
{
"operation": "Page search",
"time": "60.517",
"value": "sam68"
},
{
"operation": "Highlighting",
"time": "71.765",
"value": "KHDRBS1_sentence_6"
},
{
"operation": "Highlighting",
"time": "92.629",
"value": "KHDRBS1_sentence_12"
},
{
"operation": "Highlighting deleted",
"time": "95.939",
"value": "KHDRBS1_sentence_6"
},
{
"operation": "Highlighting",
"time": "110.884",
"value": "KHDRBS1_sentence_15"
},
{
"operation": "hyperlink",
"time": "115.436",
"value": "Metadherin"
},
{
"operation": "Now on",
"time": "116.334",
"value": "MTDH"
},
{
"operation": "Highlighting",
"time": "118.91",
"value": "MTDH_sentence_1"
},
{
"operation": "finish",
"time": "125.168",
"value": "finish"
}
]
}
|
33212
|
James McSherry's, a former Scottish football player and manager, played as a midfielder for Kilmarnock where he had 118 appearances and scored 8 goals.
|
{
"evidence": [
"Jim McSherry_sentence_0",
"Jim McSherry_sentence_1",
"Jim McSherry_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:\nJames McSherry's, a former Scottish football player and manager, played as a midfielder for Kilmarnock where he had 118 appearances and scored 8 goals.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Jim McSherry_sentence_0",
"Jim McSherry_sentence_1",
"Jim McSherry_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7647,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.358",
"value": "?search="
},
{
"operation": "Now on",
"time": "2.421",
"value": "?search="
},
{
"operation": "search",
"time": "7.516",
"value": "James McSherry"
},
{
"operation": "Now on",
"time": "8.099",
"value": "James McSherry"
},
{
"operation": "search",
"time": "44.523",
"value": "James McSherry's"
},
{
"operation": "Now on",
"time": "46.471",
"value": "undefined"
},
{
"operation": "search",
"time": "495.142",
"value": "jim mcsherry"
},
{
"operation": "Now on",
"time": "495.845",
"value": "Jim McSherry"
},
{
"operation": "Highlighting",
"time": "497.755",
"value": "Jim McSherry_sentence_0"
},
{
"operation": "Highlighting",
"time": "505.191",
"value": "Jim McSherry_sentence_2"
},
{
"operation": "Highlighting",
"time": "525.358",
"value": "Jim McSherry_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "527.56",
"value": "Jim McSherry_sentence_2"
},
{
"operation": "Highlighting",
"time": "527.735",
"value": "Jim McSherry_sentence_2"
},
{
"operation": "hyperlink",
"time": "537.077",
"value": "Kilmarnock"
},
{
"operation": "Now on",
"time": "538.053",
"value": "Kilmarnock F.C."
},
{
"operation": "search",
"time": "581.327",
"value": "1970 Kilmarnock F.C."
},
{
"operation": "Now on",
"time": "583.47",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "610.528",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "610.663",
"value": "Kilmarnock F.C."
},
{
"operation": "back-button-clicked",
"time": "611.625",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "611.628",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "611.686",
"value": "Jim McSherry"
},
{
"operation": "finish",
"time": "778.373",
"value": "finish"
}
]
}
|
exfever_6575
|
The 2013 NBA draft on June 27, 2013, was the first draft for New Orleans under their new names often referred as the Pels, a Canadian professional basketball team based in New Orleans.
|
{
"evidence": [
"2013 NBA draft",
"New Orleans Pelicans"
],
"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 2013 NBA draft on June 27, 2013, was the first draft for New Orleans under their new names often referred as the Pels, a Canadian professional basketball team based in New Orleans.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"2013 NBA draft",
"New Orleans Pelicans"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7648,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8670
|
Reds is a 1982 American epic historical drama film that stars Diane Keaton as a feminist, political activist, and journalist best known for her sympathetic coverage of Russia and the Bolsheviks during the Russian Revolution.
|
{
"evidence": [
"Reds (film)",
"Louise Bryant"
],
"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:\nReds is a 1982 American epic historical drama film that stars Diane Keaton as a feminist, political activist, and journalist best known for her sympathetic coverage of Russia and the Bolsheviks during the Russian Revolution.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Reds (film)",
"Louise Bryant"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7649,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
52111
|
Roger Sewell Bacon, a British judge who was Chief Justice of Gibraltar (1946–55) and a Justice of Appeal on the East African Court of Appeal (1955–57) who was called to the bar in 1923, had been appointed a Member of the Order of the British Empire (MBE) in 1943.
|
{
"evidence": [
"Roger Sewell Bacon_sentence_0",
"Roger Sewell Bacon_sentence_4",
"Roger Sewell Bacon_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:\nRoger Sewell Bacon, a British judge who was Chief Justice of Gibraltar (1946–55) and a Justice of Appeal on the East African Court of Appeal (1955–57) who was called to the bar in 1923, had been appointed a Member of the Order of the British Empire (MBE) in 1943.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Roger Sewell Bacon_sentence_0",
"Roger Sewell Bacon_sentence_4",
"Roger Sewell Bacon_sentence_7"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7650,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.705",
"value": "?search="
},
{
"operation": "search",
"time": "21.271",
"value": "Roger Sewell Bacon"
},
{
"operation": "Now on",
"time": "22.261",
"value": "Roger Sewell Bacon"
},
{
"operation": "Highlighting",
"time": "27.39",
"value": "Roger Sewell Bacon_sentence_0"
},
{
"operation": "Highlighting",
"time": "114.021",
"value": "Roger Sewell Bacon_sentence_4"
},
{
"operation": "Highlighting",
"time": "154.298",
"value": "Roger Sewell Bacon_sentence_7"
},
{
"operation": "finish",
"time": "194.685",
"value": "finish"
}
]
}
|
44053
|
Tim Hargreaves was an Australian rules footballer who played in the AFL, leaving Hawthorn in a trade with Geelong at the end of the 1996 season where he went on to make little impact on his new team.
|
{
"evidence": [
"Tim Hargreaves_sentence_0",
"Tim Hargreaves_sentence_4",
"Tim Hargreaves_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:\nTim Hargreaves was an Australian rules footballer who played in the AFL, leaving Hawthorn in a trade with Geelong at the end of the 1996 season where he went on to make little impact on his new team.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Tim Hargreaves_sentence_0",
"Tim Hargreaves_sentence_4",
"Tim Hargreaves_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7651,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.806",
"value": "?search="
},
{
"operation": "search",
"time": "37.062",
"value": "Tim Hargreaves"
},
{
"operation": "Now on",
"time": "38.291",
"value": "Tim Hargreaves"
},
{
"operation": "Highlighting",
"time": "39.684",
"value": "Tim Hargreaves_sentence_0"
},
{
"operation": "Highlighting",
"time": "56.397",
"value": "Tim Hargreaves_sentence_4"
},
{
"operation": "Highlighting",
"time": "66.578",
"value": "Tim Hargreaves_sentence_5"
},
{
"operation": "finish",
"time": "80.961",
"value": "finish"
}
]
}
|
68857
|
On April 16, 1881, a Royal Decree directed that a new naval base would be built in the northeastern part of Salamis Island, acquiring an area of approximately 300 hectares.
|
{
"evidence": [
"Salamis Naval Base_sentence_8",
"Salamis Naval Base_sentence_1",
"Salamis Naval Base_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:\nOn April 16, 1881, a Royal Decree directed that a new naval base would be built in the northeastern part of Salamis Island, acquiring an area of approximately 300 hectares.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Salamis Naval Base_sentence_8",
"Salamis Naval Base_sentence_1",
"Salamis Naval Base_sentence_9"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7652,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.9",
"value": "undefined"
},
{
"operation": "search",
"time": "18.744",
"value": "Contains...On April 16, 1881, a Royal Decree directed that a new naval base would be built"
},
{
"operation": "Now on",
"time": "22.836",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "27.707",
"value": "Salamis Naval Base"
},
{
"operation": "Now on",
"time": "32.22",
"value": "Salamis Naval Base"
},
{
"operation": "Page search",
"time": "37.35",
"value": "On April 16, 1881, a Royal Decree directed that a new naval base would be built"
},
{
"operation": "Highlighting",
"time": "41.125",
"value": "Salamis Naval Base_sentence_8"
},
{
"operation": "Highlighting",
"time": "58.818",
"value": "Salamis Naval Base_sentence_1"
},
{
"operation": "Highlighting",
"time": "61.707",
"value": "Salamis Naval Base_sentence_9"
},
{
"operation": "finish",
"time": "84.403",
"value": "finish"
}
]
}
|
exfever_6010
|
Zinedine Zidane's first win as a football player was as FIFA World Player of the Year in 1998.
|
{
"evidence": [
"Zinedine Zidane",
"Manager (association football)"
],
"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:\nZinedine Zidane's first win as a football player was as FIFA World Player of the Year in 1998.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Zinedine Zidane",
"Manager (association football)"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7653,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
52329
|
Travis Wammack is a member of the Memphis Music Hall of Fame, then in 2015, he was inducted into The Southern Legend Entertainment & Performing Arts Hall of Fame.
|
{
"evidence": [
"Travis Wammack_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:\nTravis Wammack is a member of the Memphis Music Hall of Fame, then in 2015, he was inducted into The Southern Legend Entertainment & Performing Arts Hall of Fame.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Travis Wammack_sentence_11"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7654,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.971",
"value": "?search="
},
{
"operation": "search",
"time": "10.105",
"value": "Travis Wammack"
},
{
"operation": "Now on",
"time": "11.02",
"value": "Travis Wammack"
},
{
"operation": "Page search",
"time": "38.914",
"value": "memphis"
},
{
"operation": "Highlighting",
"time": "51.128",
"value": "Travis Wammack_sentence_10"
},
{
"operation": "Highlighting",
"time": "57.078",
"value": "Travis Wammack_sentence_11"
},
{
"operation": "Highlighting deleted",
"time": "81.452",
"value": "Travis Wammack_sentence_10"
},
{
"operation": "finish",
"time": "92.993",
"value": "finish"
}
]
}
|
52988
|
Nezamysl was one of seven mythical princes born to Premysl the Plloughman, who was a free peasant who married Libuse and becamse prince of the Bohemian Czechs.
|
{
"evidence": [
"Nezamysl_sentence_0",
"Přemysl the Ploughman_sentence_3",
"Přemysl the Ploughman_sentence_7"
],
"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:\nNezamysl was one of seven mythical princes born to Premysl the Plloughman, who was a free peasant who married Libuse and becamse prince of the Bohemian Czechs.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Nezamysl_sentence_0",
"Přemysl the Ploughman_sentence_3",
"Přemysl the Ploughman_sentence_7"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7655,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.772",
"value": "?search="
},
{
"operation": "search",
"time": "14.32",
"value": "Nezamysl"
},
{
"operation": "Now on",
"time": "14.973",
"value": "Nezamysl"
},
{
"operation": "Highlighting",
"time": "53.371",
"value": "Nezamysl_sentence_0"
},
{
"operation": "hyperlink",
"time": "112.884",
"value": "Přemysl the Ploughman"
},
{
"operation": "Now on",
"time": "113.55",
"value": "Přemysl the Ploughman"
},
{
"operation": "hyperlink",
"time": "165.842",
"value": "Přemyslid dynasty"
},
{
"operation": "Now on",
"time": "169.541",
"value": "Přemyslid dynasty"
},
{
"operation": "Highlighting",
"time": "183.235",
"value": "Přemyslid dynasty_sentence_5"
},
{
"operation": "Highlighting",
"time": "183.236",
"value": "Přemyslid dynasty_title"
},
{
"operation": "Highlighting",
"time": "183.237",
"value": "Přemyslid dynasty_section_0"
},
{
"operation": "Highlighting deleted",
"time": "186.303",
"value": "Přemyslid dynasty_sentence_5"
},
{
"operation": "Highlighting deleted",
"time": "186.305",
"value": "Přemyslid dynasty_title"
},
{
"operation": "Highlighting deleted",
"time": "186.311",
"value": "Přemyslid dynasty_section_0"
},
{
"operation": "back-button-clicked",
"time": "189.473",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "190.086",
"value": "Přemysl the Ploughman"
},
{
"operation": "back-button-clicked",
"time": "232.169",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "232.176",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "232.237",
"value": "Nezamysl"
},
{
"operation": "hyperlink",
"time": "274.053",
"value": "Přemysl the Ploughman"
},
{
"operation": "Now on",
"time": "275.055",
"value": "Přemysl, the Ploughman"
},
{
"operation": "Highlighting",
"time": "278.461",
"value": "Přemysl the Ploughman_sentence_3"
},
{
"operation": "Highlighting",
"time": "289.733",
"value": "Přemysl the Ploughman_sentence_7"
},
{
"operation": "finish",
"time": "298.958",
"value": "finish"
}
]
}
|
94664
|
Travis Wammack is a member of the Memphis Music Hall of Fame, then in 2005, he was inducted into The Southern Legend Entertainment & Performing Arts Hall of Fame.
|
{
"evidence": [
"Travis Wammack_sentence_10",
"Travis Wammack_sentence_11"
],
"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:\nTravis Wammack is a member of the Memphis Music Hall of Fame, then in 2005, he was inducted into The Southern Legend Entertainment & Performing Arts Hall of Fame.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Travis Wammack_sentence_10",
"Travis Wammack_sentence_11"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7656,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.966",
"value": "?search="
},
{
"operation": "search",
"time": "15.033",
"value": "Travis Wammack"
},
{
"operation": "Now on",
"time": "15.924",
"value": "Travis Wammack"
},
{
"operation": "Highlighting",
"time": "28.875",
"value": "Travis Wammack_sentence_10"
},
{
"operation": "Highlighting",
"time": "35.627",
"value": "Travis Wammack_sentence_11"
},
{
"operation": "finish",
"time": "58.511",
"value": "finish"
}
]
}
|
93076
|
The Daba Mountains are home to dawn redwood which are considered living fossils, they resemble organisms known only from the fossil records.
|
{
"evidence": [
"Daba Mountains_sentence_18",
"Metasequoia glyptostroboides_sentence_4",
"Living fossil_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:\nThe Daba Mountains are home to dawn redwood which are considered living fossils, they resemble organisms known only from the fossil records.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Daba Mountains_sentence_18",
"Metasequoia glyptostroboides_sentence_4",
"Living fossil_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7657,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.41",
"value": "?search="
},
{
"operation": "search",
"time": "31.111",
"value": "Daba Mountains"
},
{
"operation": "Now on",
"time": "32.225",
"value": "Daba Mountains"
},
{
"operation": "Page search",
"time": "40.818",
"value": "dawn"
},
{
"operation": "Highlighting",
"time": "52.376",
"value": "Daba Mountains_sentence_18"
},
{
"operation": "hyperlink",
"time": "77.851",
"value": "Dawn Redwood"
},
{
"operation": "Now on",
"time": "79.055",
"value": "Metasequoia glyptostroboides"
},
{
"operation": "Page search",
"time": "107.979",
"value": "foss"
},
{
"operation": "Highlighting",
"time": "122.509",
"value": "Metasequoia glyptostroboides_sentence_4"
},
{
"operation": "hyperlink",
"time": "127.77",
"value": "living fossil"
},
{
"operation": "Now on",
"time": "128.887",
"value": "Living fossil"
},
{
"operation": "Highlighting",
"time": "130.385",
"value": "Living fossil_sentence_2"
},
{
"operation": "finish",
"time": "135.64",
"value": "finish"
}
]
}
|
exfever_3116
|
The New York Giants were the champion of the National Football Conference in 2009.
|
{
"evidence": [
"Super Bowl XLVI",
"National Football League"
],
"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 New York Giants were the champion of the National Football Conference in 2009.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Super Bowl XLVI",
"National Football League"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7658,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7110
|
Zooey Deschanel is known for her deadpan roles in comedies, such as in the 2002 American comedy-drama film directed by a Puerto Rican director known for his independent film Chuck and Buck.
|
{
"evidence": [
"Zooey Deschanel",
"The Good Girl",
"Miguel Arteta"
],
"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:\nZooey Deschanel is known for her deadpan roles in comedies, such as in the 2002 American comedy-drama film directed by a Puerto Rican director known for his independent film Chuck and Buck.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Zooey Deschanel",
"The Good Girl",
"Miguel Arteta"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7659,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
77821
|
When Rappers Attack was the first studio album released by Tame One (which contained a hit single with the same name) in 2003 which made it to number 74 on the "100 Best Indie Hip-Hop Records of All Time" list.
|
{
"evidence": [
"When Rappers Attack_sentence_0",
"When Rappers Attack_sentence_1",
"When Rappers Attack_sentence_4"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nWhen Rappers Attack was the first studio album released by Tame One (which contained a hit single with the same name) in 2003 which made it to number 74 on the \"100 Best Indie Hip-Hop Records of All Time\" list.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"When Rappers Attack_sentence_0",
"When Rappers Attack_sentence_1",
"When Rappers Attack_sentence_4"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7660,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.093",
"value": "?search="
},
{
"operation": "search",
"time": "14.154",
"value": "When Rappers Attack"
},
{
"operation": "Now on",
"time": "15.059",
"value": "When Rappers Attack"
},
{
"operation": "Highlighting",
"time": "21.922",
"value": "When Rappers Attack_sentence_0"
},
{
"operation": "Highlighting",
"time": "43.818",
"value": "When Rappers Attack_sentence_1"
},
{
"operation": "Highlighting",
"time": "49.258",
"value": "When Rappers Attack_sentence_4"
},
{
"operation": "finish",
"time": "54.206",
"value": "finish"
}
]
}
|
exfever_2909
|
Trolls is a 2017 American computer-animated musical comedy film produced by DreamWorks Animation featuring the voice of one of the world's best-selling music artists, with sales of over 88 million records.
|
{
"evidence": [
"Trolls (film)",
"Justin Timberlake"
],
"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:\nTrolls is a 2017 American computer-animated musical comedy film produced by DreamWorks Animation featuring the voice of one of the world's best-selling music artists, with sales of over 88 million records.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Trolls (film)",
"Justin Timberlake"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7661,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2208
|
The Big Trail is a 1930 American pre-Code Southern early widescreen film directed by an American film director who is known for directing The Roaring Twenties starring an actor known for his consistently energetic performances, distinctive vocal style, and deadpan comic timing.
|
{
"evidence": [
"The Big Trail",
"Raoul Walsh",
"James Cagney"
],
"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 Big Trail is a 1930 American pre-Code Southern early widescreen film directed by an American film director who is known for directing The Roaring Twenties starring an actor known for his consistently energetic performances, distinctive vocal style, and deadpan comic timing.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Big Trail",
"Raoul Walsh",
"James Cagney"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7662,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2459
|
The Divergent Series was distributed by an Ecuadorean film production and distribution company.
|
{
"evidence": [
"The Divergent Series",
"Summit Entertainment"
],
"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 Divergent Series was distributed by an Ecuadorean film production and distribution company.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Divergent Series",
"Summit Entertainment"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7663,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
195
|
Hart to Hart had 24 episodes featured for Season 3 from 1981 to 1982.
|
{
"evidence": [
"Stefanie Powers_sentence_52"
],
"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:\nHart to Hart had 24 episodes featured for Season 3 from 1981 to 1982.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Stefanie Powers_sentence_52"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7664,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "search",
"time": "0",
"value": "Hart to Hart"
},
{
"operation": "hyperlink",
"time": "12.709",
"value": "Stefanie Powers"
},
{
"operation": "Highlighting",
"time": "88.038",
"value": "Stefanie Powers_sentence_52"
},
{
"operation": "search",
"time": "133.827",
"value": "Season 3 Hart to Hart"
},
{
"operation": "finish",
"time": "166.092",
"value": "finish"
}
]
}
|
70668
|
After the death of Nero, when the chaos of the Year of the Four Emperors had settled, Africanus was prosecuted for his role in the deaths of the brothers Sulpicii Scribonii; he instead implicated Quintus Vibius Crispus as also involved in the matter, resulting to his promotion to the Curia Julia.
|
{
"evidence": [
"Gaius Paccius Africanus_sentence_6",
"Gaius Paccius Africanus_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:\nAfter the death of Nero, when the chaos of the Year of the Four Emperors had settled, Africanus was prosecuted for his role in the deaths of the brothers Sulpicii Scribonii; he instead implicated Quintus Vibius Crispus as also involved in the matter, resulting to his promotion to the Curia Julia.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Gaius Paccius Africanus_sentence_6",
"Gaius Paccius Africanus_sentence_7"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7665,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.224",
"value": "?search="
},
{
"operation": "search",
"time": "4.661",
"value": "Nero"
},
{
"operation": "Now on",
"time": "5.421",
"value": "Nero"
},
{
"operation": "Highlighting",
"time": "73.546",
"value": "Nero_sentence_18"
},
{
"operation": "Highlighting",
"time": "73.549",
"value": "Nero_title"
},
{
"operation": "Highlighting",
"time": "74.986",
"value": "Nero_sentence_19"
},
{
"operation": "Highlighting deleted",
"time": "87.552",
"value": "Nero_sentence_18"
},
{
"operation": "Highlighting deleted",
"time": "88.61",
"value": "Nero_sentence_19"
},
{
"operation": "Highlighting deleted",
"time": "88.611",
"value": "Nero_title"
},
{
"operation": "hyperlink",
"time": "89.645",
"value": "Year of the Four Emperors"
},
{
"operation": "Now on",
"time": "90.469",
"value": "Year of the Four Emperors"
},
{
"operation": "Highlighting",
"time": "104.382",
"value": "Year of the Four Emperors_sentence_1"
},
{
"operation": "Highlighting",
"time": "104.383",
"value": "Year of the Four Emperors_title"
},
{
"operation": "Highlighting",
"time": "107.836",
"value": "Year of the Four Emperors_sentence_2"
},
{
"operation": "search",
"time": "137.098",
"value": "Sulpicii Scibonii"
},
{
"operation": "Now on",
"time": "138.765",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "141.013",
"value": "sulpicius scionti"
},
{
"operation": "Now on",
"time": "141.549",
"value": "undefined"
},
{
"operation": "search",
"time": "147.199",
"value": "Africanus"
},
{
"operation": "Now on",
"time": "147.853",
"value": "Africanus"
},
{
"operation": "Highlighting",
"time": "163.074",
"value": "Africanus_item_0_3"
},
{
"operation": "Highlighting",
"time": "163.075",
"value": "Africanus_title"
},
{
"operation": "Highlighting deleted",
"time": "164.48",
"value": "Africanus_item_0_3"
},
{
"operation": "Highlighting deleted",
"time": "164.481",
"value": "Africanus_title"
},
{
"operation": "hyperlink",
"time": "165.515",
"value": "Julius Africanus"
},
{
"operation": "Now on",
"time": "166.172",
"value": "Julius Africanus"
},
{
"operation": "hyperlink",
"time": "188.601",
"value": "Africanus"
},
{
"operation": "Now on",
"time": "195.807",
"value": "Africanus (disambiguation)"
},
{
"operation": "hyperlink",
"time": "216.444",
"value": "Julius Africanus"
},
{
"operation": "Now on",
"time": "217.184",
"value": "Julius Africanus"
},
{
"operation": "search",
"time": "241.532",
"value": "Quinton Vibius Crispus"
},
{
"operation": "Now on",
"time": "242.908",
"value": "undefined"
},
{
"operation": "search",
"time": "253.378",
"value": "Curia Julia"
},
{
"operation": "Now on",
"time": "254.835",
"value": "Curia Julia"
},
{
"operation": "search",
"time": "290.394",
"value": "Quintus Vibius Secundus"
},
{
"operation": "Now on",
"time": "291.213",
"value": "Quintus Vibius Secundus"
},
{
"operation": "search",
"time": "311.226",
"value": "Africanus prosecuted"
},
{
"operation": "Now on",
"time": "312.664",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "317.37",
"value": "Gaius Paccius Africanus"
},
{
"operation": "Now on",
"time": "318.037",
"value": "Gaius Paccius Africanus"
},
{
"operation": "Highlighting",
"time": "331.625",
"value": "Gaius Paccius Africanus_sentence_6"
},
{
"operation": "Highlighting",
"time": "350.344",
"value": "Gaius Paccius Africanus_sentence_7"
},
{
"operation": "Now on",
"time": "368.926",
"value": "Gaius Paccius Africanus"
},
{
"operation": "finish",
"time": "399.181",
"value": "finish"
}
]
}
|
15425
|
Gajularega is a census town in Vizianagaram district in the state of Uttar Pradesh, India with a total population of 13,078 as of 2001.
|
{
"evidence": [
"Gajularega_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:\nGajularega is a census town in Vizianagaram district in the state of Uttar Pradesh, India with a total population of 13,078 as of 2001.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Gajularega_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7666,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.37",
"value": "?search="
},
{
"operation": "search",
"time": "10.645",
"value": "Gajularega"
},
{
"operation": "Now on",
"time": "11.426",
"value": "Gajularega"
},
{
"operation": "Highlighting",
"time": "17.606",
"value": "Gajularega_sentence_0"
},
{
"operation": "finish",
"time": "87.728",
"value": "finish"
}
]
}
|
90444
|
The characters Oko and Seiji, a clever man who is known to solve mysteries, run a lending shop in Edo, one of the largest cities iin the world under the Tokugaya, renting out old items.
|
{
"evidence": [
"Tsukumogami Kashimasu_sentence_4",
"Tsukumogami Kashimasu_sentence_5",
"Tsukumogami Kashimasu_sentence_6",
"Edo_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:\nThe characters Oko and Seiji, a clever man who is known to solve mysteries, run a lending shop in Edo, one of the largest cities iin the world under the Tokugaya, renting out old items.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Tsukumogami Kashimasu_sentence_4",
"Tsukumogami Kashimasu_sentence_5",
"Tsukumogami Kashimasu_sentence_6",
"Edo_sentence_7"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7667,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.424",
"value": "?search="
},
{
"operation": "search",
"time": "12.787",
"value": "Oko and Seiji"
},
{
"operation": "Now on",
"time": "15.022",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "57.95",
"value": "Tsukumogami Kashimasu"
},
{
"operation": "Now on",
"time": "58.826",
"value": "Tsukumogami Kashimasu"
},
{
"operation": "Page search",
"time": "61.296",
"value": "Oko and Seiji"
},
{
"operation": "Highlighting",
"time": "77.186",
"value": "Tsukumogami Kashimasu_sentence_4"
},
{
"operation": "Highlighting",
"time": "80.716",
"value": "Tsukumogami Kashimasu_sentence_6"
},
{
"operation": "hyperlink",
"time": "93.18",
"value": "Edo"
},
{
"operation": "Now on",
"time": "94.282",
"value": "Edo"
},
{
"operation": "back-button-clicked",
"time": "97.902",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "97.97",
"value": "Tsukumogami Kashimasu"
},
{
"operation": "Highlighting deleted",
"time": "105.258",
"value": "Tsukumogami Kashimasu_sentence_6"
},
{
"operation": "Highlighting",
"time": "106.268",
"value": "Tsukumogami Kashimasu_sentence_5"
},
{
"operation": "Highlighting",
"time": "107.262",
"value": "Tsukumogami Kashimasu_sentence_6"
},
{
"operation": "hyperlink",
"time": "109.877",
"value": "Edo"
},
{
"operation": "Now on",
"time": "110.503",
"value": "Edo"
},
{
"operation": "Highlighting",
"time": "116.318",
"value": "Edo_sentence_7"
},
{
"operation": "finish",
"time": "123.531",
"value": "finish"
}
]
}
|
17415
|
Stijn Spierings was born in Alkmaar (a city and municipality in the Netherlands, located in the province of North Holland), and started his career with AZ and played on several occasions in his first season.
|
{
"evidence": [
"Stijn Spierings_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:\nStijn Spierings was born in Alkmaar (a city and municipality in the Netherlands, located in the province of North Holland), and started his career with AZ and played on several occasions in his first season.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Stijn Spierings_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7668,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.767",
"value": "?search="
},
{
"operation": "search",
"time": "8.361",
"value": "Stijn Spierings"
},
{
"operation": "Now on",
"time": "9.253",
"value": "Stijn Spierings"
},
{
"operation": "Highlighting",
"time": "13.774",
"value": "Stijn Spierings_cell_0_3_1"
},
{
"operation": "Highlighting",
"time": "13.795",
"value": "Stijn Spierings_header_cell_0_3_0"
},
{
"operation": "Highlighting",
"time": "13.795",
"value": "Stijn Spierings_header_cell_0_0_0"
},
{
"operation": "hyperlink",
"time": "17.201",
"value": "Alkmaar"
},
{
"operation": "Now on",
"time": "18.279",
"value": "Alkmaar"
},
{
"operation": "Highlighting",
"time": "19.439",
"value": "Alkmaar_sentence_1"
},
{
"operation": "Highlighting",
"time": "19.441",
"value": "Alkmaar_title"
},
{
"operation": "back-button-clicked",
"time": "20.722",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "20.845",
"value": "Stijn Spierings"
},
{
"operation": "hyperlink",
"time": "26.049",
"value": "Alkmaar"
},
{
"operation": "Now on",
"time": "26.547",
"value": "Alkmaar"
},
{
"operation": "back-button-clicked",
"time": "30.628",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "30.745",
"value": "Stijn Spierings"
},
{
"operation": "Highlighting",
"time": "34.025",
"value": "Stijn Spierings_sentence_1"
},
{
"operation": "Highlighting",
"time": "40.589",
"value": "Stijn Spierings_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "43.925",
"value": "Stijn Spierings_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "45.731",
"value": "Stijn Spierings_cell_0_3_1"
},
{
"operation": "Highlighting deleted",
"time": "45.75",
"value": "Stijn Spierings_header_cell_0_3_0"
},
{
"operation": "Highlighting deleted",
"time": "45.751",
"value": "Stijn Spierings_header_cell_0_0_0"
},
{
"operation": "finish",
"time": "65.774",
"value": "finish"
}
]
}
|
85952
|
Ben Swift learned how to drive a motorcycle at the age of 16.
|
{
"evidence": [
"Ben Swift_sentence_3"
],
"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:\nBen Swift learned how to drive a motorcycle at the age of 16.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Ben Swift_sentence_3"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7669,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.778",
"value": "?search="
},
{
"operation": "search",
"time": "10.431",
"value": "Ben Swift"
},
{
"operation": "Now on",
"time": "11.42",
"value": "Ben Swift"
},
{
"operation": "Highlighting",
"time": "24.173",
"value": "Ben Swift_sentence_3"
},
{
"operation": "Highlighting",
"time": "82.438",
"value": "Ben Swift_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "88.186",
"value": "Ben Swift_sentence_0"
},
{
"operation": "Page search",
"time": "106.881",
"value": "motorcycle"
},
{
"operation": "finish",
"time": "117.57",
"value": "finish"
}
]
}
|
exfever_4260
|
Led Zeppelin group consisted of vocalist Robert Plant, guitarist Jimmy Page, drummer John Bonham, and the bassist/keyboardist who collaborated with musicians including an American musician best known as the founder of the rock band Queens of the Industrial Age.
|
{
"evidence": [
"Led Zeppelin",
"John Paul Jones (musician)",
"Josh Homme"
],
"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:\nLed Zeppelin group consisted of vocalist Robert Plant, guitarist Jimmy Page, drummer John Bonham, and the bassist/keyboardist who collaborated with musicians including an American musician best known as the founder of the rock band Queens of the Industrial Age.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Led Zeppelin",
"John Paul Jones (musician)",
"Josh Homme"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7670,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7945
|
Aubrey Drake Graham is a Canadian rapper, singer, songwriter, and actor who released his debut studio album on June 15, 2011, by an American record label founded by rapper Lil Wayne.
|
{
"evidence": [
"Drake (musician)",
"Thank Me Later",
"Young Money Entertainment"
],
"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:\nAubrey Drake Graham is a Canadian rapper, singer, songwriter, and actor who released his debut studio album on June 15, 2011, by an American record label founded by rapper Lil Wayne.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Drake (musician)",
"Thank Me Later",
"Young Money Entertainment"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7671,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3484
|
Chris Wood made a surprise reappearance as Mon-El in Season 5's "It's a Super Life," the series' milestone 100th episode.
|
{
"evidence": [
"James Bond",
"John Gardner (British writer)"
],
"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:\nChris Wood made a surprise reappearance as Mon-El in Season 5's \"It's a Super Life,\" the series' milestone 100th episode.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"James Bond",
"John Gardner (British writer)"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7672,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
88704
|
The Campeones Cup's 2020 edition was canceled due to the COVID-19 pandemic, which started in December 2019.
|
{
"evidence": [
"COVID-19 pandemic_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:\nThe Campeones Cup's 2020 edition was canceled due to the COVID-19 pandemic, which started in December 2019.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"COVID-19 pandemic_sentence_1"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7673,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.907",
"value": "?search="
},
{
"operation": "search",
"time": "155.852",
"value": "Campeones Cup"
},
{
"operation": "Now on",
"time": "157.303",
"value": "Campeones Cup"
},
{
"operation": "Highlighting",
"time": "202.266",
"value": "Campeones Cup_sentence_10"
},
{
"operation": "Highlighting",
"time": "202.267",
"value": "Campeones Cup_title"
},
{
"operation": "Highlighting",
"time": "202.268",
"value": "Campeones Cup_section_1"
},
{
"operation": "hyperlink",
"time": "210.385",
"value": "COVID-19 pandemic"
},
{
"operation": "Now on",
"time": "214.101",
"value": "COVID-19 pandemic"
},
{
"operation": "Highlighting",
"time": "227.978",
"value": "COVID-19 pandemic_sentence_1"
},
{
"operation": "back-button-clicked",
"time": "261.234",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "261.447",
"value": "Campeones Cup"
},
{
"operation": "Highlighting deleted",
"time": "262.938",
"value": "Campeones Cup_sentence_10"
},
{
"operation": "Highlighting deleted",
"time": "262.939",
"value": "Campeones Cup_title"
},
{
"operation": "Highlighting deleted",
"time": "262.941",
"value": "Campeones Cup_section_1"
},
{
"operation": "finish",
"time": "279.65",
"value": "finish"
}
]
}
|
84498
|
Alan Horsley was born on 13 May 1936, his first post was a curate in Daventry but did not follow any religion.
|
{
"evidence": [
"Alan Horsley_sentence_0",
"Anglicanism_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:\nAlan Horsley was born on 13 May 1936, his first post was a curate in Daventry but did not follow any religion.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Alan Horsley_sentence_0",
"Anglicanism_sentence_5"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7674,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.68",
"value": "?search="
},
{
"operation": "search",
"time": "14.777",
"value": "Alan Horsley"
},
{
"operation": "Now on",
"time": "15.511",
"value": "Alan Horsley"
},
{
"operation": "Highlighting",
"time": "19.623",
"value": "Alan Horsley_sentence_1"
},
{
"operation": "Highlighting",
"time": "62.07",
"value": "Alan Horsley_sentence_2"
},
{
"operation": "hyperlink",
"time": "112.422",
"value": "Provost of St Andrew's Cathedral, Inverness"
},
{
"operation": "Now on",
"time": "112.889",
"value": "Inverness Cathedral"
},
{
"operation": "back-button-clicked",
"time": "119.702",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "119.753",
"value": "Alan Horsley"
},
{
"operation": "Highlighting",
"time": "128.115",
"value": "Alan Horsley_sentence_0"
},
{
"operation": "hyperlink",
"time": "129.684",
"value": "Anglican"
},
{
"operation": "Now on",
"time": "132.103",
"value": "Anglicanism"
},
{
"operation": "Highlighting",
"time": "147.877",
"value": "Anglicanism_sentence_5"
},
{
"operation": "back-button-clicked",
"time": "187.558",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "187.617",
"value": "Alan Horsley"
},
{
"operation": "Highlighting deleted",
"time": "192.277",
"value": "Alan Horsley_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "194.198",
"value": "Alan Horsley_sentence_2"
},
{
"operation": "finish",
"time": "365.493",
"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": 7675,
"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"
}
]
}
|
23311
|
Christian Democratic Party politician Odd Holten (08/28/1940-- ) was elected to the Norwegian Parliament from Østfold in 1989.
|
{
"evidence": [
"Odd Holten_sentence_0",
"Odd Holten_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:\nChristian Democratic Party politician Odd Holten (08/28/1940-- ) was elected to the Norwegian Parliament from Østfold in 1989.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Odd Holten_sentence_0",
"Odd Holten_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7676,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.022",
"value": "?search="
},
{
"operation": "search",
"time": "17.529",
"value": "Odd Holten"
},
{
"operation": "Now on",
"time": "18.408",
"value": "Odd Holten"
},
{
"operation": "Highlighting",
"time": "25.546",
"value": "Odd Holten_sentence_0"
},
{
"operation": "Highlighting",
"time": "34.106",
"value": "Odd Holten_sentence_1"
},
{
"operation": "finish",
"time": "53.942",
"value": "finish"
}
]
}
|
exfever_3402
|
Slow Jamz was the first hit song by Kanye West.
|
{
"evidence": [
"Kanye West",
"Roc-A-Fella Records"
],
"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:\nSlow Jamz was the first hit song by Kanye West.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Kanye West",
"Roc-A-Fella Records"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7677,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5771
|
Hugo Weaving was an English actor whose first major role was as a cricketer and captain who played 22 Test matches for America.
|
{
"evidence": [
"Hugo Weaving",
"Douglas Jardine"
],
"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:\nHugo Weaving was an English actor whose first major role was as a cricketer and captain who played 22 Test matches for America.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Hugo Weaving",
"Douglas Jardine"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7678,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2476
|
24 is a 2016 Indian Tamil-language science fiction action film with the co-star of an American actress and singer who works primarily in Malayalam, Telugu, Tamil, and Kannada films.
|
{
"evidence": [
"24 (2016 film)",
"Nithya Menen"
],
"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:\n24 is a 2016 Indian Tamil-language science fiction action film with the co-star of an American actress and singer who works primarily in Malayalam, Telugu, Tamil, and Kannada films.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"24 (2016 film)",
"Nithya Menen"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7679,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3902
|
Cleopatra VII Philopator was Queen of the Ptolemaic Kingdom of Egypt from 51 to 30 BC and was a member of the Lagid dynasty, named after a Macedonian courtier who married Arsinoe of the Argead dynasty.
|
{
"evidence": [
"Cleopatra",
"Ptolemaic dynasty",
"Lagus"
],
"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:\nCleopatra VII Philopator was Queen of the Ptolemaic Kingdom of Egypt from 51 to 30 BC and was a member of the Lagid dynasty, named after a Macedonian courtier who married Arsinoe of the Argead dynasty.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Cleopatra",
"Ptolemaic dynasty",
"Lagus"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7680,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
94302
|
The 1937 Michigan State Normal Hurons football team represented Michigan State Normal College which is known as Eastern Michigan University and is a public research university in Ypsilanti, Michigan.
|
{
"evidence": [
"1937 Michigan State Normal Hurons football team_sentence_0",
"Eastern Michigan University_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:\nThe 1937 Michigan State Normal Hurons football team represented Michigan State Normal College which is known as Eastern Michigan University and is a public research university in Ypsilanti, Michigan.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"1937 Michigan State Normal Hurons football team_sentence_0",
"Eastern Michigan University_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7681,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.032",
"value": "?search="
},
{
"operation": "search",
"time": "7.434",
"value": "1937 Michigan State Normal Hurons football team"
},
{
"operation": "Now on",
"time": "8.359",
"value": "1937 Michigan State Normal Hurons football team"
},
{
"operation": "Highlighting",
"time": "12.518",
"value": "1937 Michigan State Normal Hurons football team_sentence_0"
},
{
"operation": "hyperlink",
"time": "24.912",
"value": "Michigan State Normal College"
},
{
"operation": "Now on",
"time": "26.018",
"value": "Eastern Michigan University"
},
{
"operation": "Highlighting",
"time": "34.268",
"value": "Eastern Michigan University_sentence_0"
},
{
"operation": "finish",
"time": "40.912",
"value": "finish"
}
]
}
|
94070
|
Nebraska Cornhuskers won four times and lost three times during the games they participated in.
|
{
"evidence": [
"Nebraska Cornhuskers_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:\nNebraska Cornhuskers won four times and lost three times during the games they participated in.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Nebraska Cornhuskers_sentence_2"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7682,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.115",
"value": "?search="
},
{
"operation": "search",
"time": "6.912",
"value": "Nebraska Cornhuskers"
},
{
"operation": "Now on",
"time": "9.041",
"value": "Nebraska Cornhuskers"
},
{
"operation": "Page search",
"time": "18.109",
"value": "won"
},
{
"operation": "Page search",
"time": "35.924",
"value": "won4-3"
},
{
"operation": "Page search",
"time": "39.298",
"value": "4-3"
},
{
"operation": "Page search",
"time": "68.789",
"value": "participate"
},
{
"operation": "Highlighting",
"time": "77.58",
"value": "Nebraska Cornhuskers_sentence_2"
},
{
"operation": "finish",
"time": "84.674",
"value": "finish"
}
]
}
|
14403
|
Martin was baptized in Olney, Buckinghamshire, England on May 30, 1621; then on August 11, 1646 at Salisbury, she married a widower named George Martin.
|
{
"evidence": [
"Susannah Martin_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:\nMartin was baptized in Olney, Buckinghamshire, England on May 30, 1621; then on August 11, 1646 at Salisbury, she married a widower named George Martin.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Susannah Martin_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7683,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.6",
"value": "?search="
},
{
"operation": "search",
"time": "152.893",
"value": "george martin"
},
{
"operation": "Now on",
"time": "154.945",
"value": "George Martin"
},
{
"operation": "search",
"time": "212.723",
"value": "Susannah Martin"
},
{
"operation": "Now on",
"time": "213.675",
"value": "Susannah Martin"
},
{
"operation": "Highlighting",
"time": "228.901",
"value": "Susannah Martin_sentence_4"
},
{
"operation": "finish",
"time": "246.822",
"value": "finish"
}
]
}
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.