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 |
|---|---|---|---|---|---|---|---|---|
52849
|
Liver X receptor alpha is one of three liver X receptors, which activate thyroid hormones and play a role in regulating the expression of target genes.
|
{
"evidence": [
"Liver X receptor alpha_sentence_4",
"Liver X receptor_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:\nLiver X receptor alpha is one of three liver X receptors, which activate thyroid hormones and play a role in regulating the expression of target genes.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Liver X receptor alpha_sentence_4",
"Liver X receptor_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 584,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.235",
"value": "?search="
},
{
"operation": "search",
"time": "9.712",
"value": "Liver X receptor alpha"
},
{
"operation": "Now on",
"time": "10.627",
"value": "Liver X receptor alpha"
},
{
"operation": "Page search",
"time": "83.24",
"value": "receptor"
},
{
"operation": "Page search",
"time": "94.317",
"value": "liver x receptor"
},
{
"operation": "Page search",
"time": "95.464",
"value": "liver x receptor"
},
{
"operation": "Page search",
"time": "119.467",
"value": "thyroid"
},
{
"operation": "hyperlink",
"time": "126.646",
"value": "liver X receptors"
},
{
"operation": "Now on",
"time": "127.626",
"value": "Liver X receptor"
},
{
"operation": "Page search",
"time": "264.11",
"value": "alpha"
},
{
"operation": "Page search",
"time": "264.826",
"value": "alpha"
},
{
"operation": "Page search",
"time": "289.935",
"value": "thyroid"
},
{
"operation": "back-button-clicked",
"time": "301.073",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "301.139",
"value": "Liver X receptor alpha"
},
{
"operation": "Page search",
"time": "306.408",
"value": "thyroid"
},
{
"operation": "Highlighting",
"time": "309.631",
"value": "Liver X receptor alpha_sentence_4"
},
{
"operation": "hyperlink",
"time": "367.541",
"value": "liver X receptors"
},
{
"operation": "Now on",
"time": "369.21",
"value": "Liver X receptor"
},
{
"operation": "Highlighting",
"time": "372.211",
"value": "Liver X receptor_sentence_0"
},
{
"operation": "finish",
"time": "426.3",
"value": "finish"
}
]
}
|
exfever_6686
|
The human condition is a broad topic that has been and continues to be pondered and analyzed from many perspectives, including historical research and documentation.
|
{
"evidence": [
"Human condition",
"History"
],
"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 human condition is a broad topic that has been and continues to be pondered and analyzed from many perspectives, including historical research and documentation.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Human condition",
"History"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 585,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
86862
|
Art Hirahara is a jazz pianist and composer who had lessons at Oberlin with pianist Neal Creque.
|
{
"evidence": [
"Art Hirahara_sentence_0",
"Art Hirahara_sentence_2",
"Art Hirahara_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:\nArt Hirahara is a jazz pianist and composer who had lessons at Oberlin with pianist Neal Creque.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Art Hirahara_sentence_0",
"Art Hirahara_sentence_2",
"Art Hirahara_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 586,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.96",
"value": "?search="
},
{
"operation": "search",
"time": "9.931",
"value": "Art Hirahara"
},
{
"operation": "Now on",
"time": "10.855",
"value": "Art Hirahara"
},
{
"operation": "Highlighting",
"time": "14.846",
"value": "Art Hirahara_sentence_0"
},
{
"operation": "Page search",
"time": "24.504",
"value": "Oberlin"
},
{
"operation": "Highlighting",
"time": "28.583",
"value": "Art Hirahara_sentence_2"
},
{
"operation": "Highlighting",
"time": "30.261",
"value": "Art Hirahara_sentence_3"
},
{
"operation": "finish",
"time": "57.102",
"value": "finish"
}
]
}
|
88221
|
Pamelyn Ferdin is an American animal rights activist and a former child actress who played a role in the films, The Beguiled and The Mephisto Waltz in 1973.
|
{
"evidence": [
"Pamelyn Ferdin_sentence_16"
],
"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:\nPamelyn Ferdin is an American animal rights activist and a former child actress who played a role in the films, The Beguiled and The Mephisto Waltz in 1973.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Pamelyn Ferdin_sentence_16"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 587,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.361",
"value": "?search="
},
{
"operation": "search",
"time": "22.996",
"value": "Pamelyn Ferdin"
},
{
"operation": "Now on",
"time": "25.274",
"value": "Pamelyn Ferdin"
},
{
"operation": "Highlighting",
"time": "29.071",
"value": "Pamelyn Ferdin_sentence_0"
},
{
"operation": "Highlighting",
"time": "50.817",
"value": "Pamelyn Ferdin_sentence_3"
},
{
"operation": "Highlighting",
"time": "107.154",
"value": "Pamelyn Ferdin_sentence_16"
},
{
"operation": "Highlighting deleted",
"time": "126.489",
"value": "Pamelyn Ferdin_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "127.839",
"value": "Pamelyn Ferdin_sentence_0"
},
{
"operation": "finish",
"time": "142.583",
"value": "finish"
}
]
}
|
27025
|
The routes within the scope of this article spring eastward from Salisbury, which was reached from Waterloo station in London; then from Salisbury, the main line continued broadly west, passing no major population centre until reaching Exeter.
|
{
"evidence": [
"Southern Railway routes west of Salisbury_sentence_2"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe routes within the scope of this article spring eastward from Salisbury, which was reached from Waterloo station in London; then from Salisbury, the main line continued broadly west, passing no major population centre until reaching Exeter.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Southern Railway routes west of Salisbury_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 588,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.074",
"value": "?search="
},
{
"operation": "search",
"time": "40.471",
"value": "Contains...within the scope of this article spring eastward from Salisbury"
},
{
"operation": "Now on",
"time": "42.135",
"value": "undefined"
},
{
"operation": "search",
"time": "57.907",
"value": "Contains...spring eastward from Salisbury"
},
{
"operation": "Now on",
"time": "58.855",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "81.813",
"value": "Wilts, Somerset and Weymouth Railway"
},
{
"operation": "Now on",
"time": "82.66",
"value": "Wilts, Somerset and Weymouth Railway"
},
{
"operation": "Page search",
"time": "108.064",
"value": "spring eastward from Salisbury"
},
{
"operation": "Page search",
"time": "189.817",
"value": "MAIN LINE CONTINUED"
},
{
"operation": "back-button-clicked",
"time": "224.973",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "225.636",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "272.002",
"value": "Tornado outbreak of April 14–16, 2011"
},
{
"operation": "Now on",
"time": "272.97",
"value": "Tornado outbreak of April 14–16, 2011"
},
{
"operation": "back-button-clicked",
"time": "284.426",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "285.174",
"value": "undefined"
},
{
"operation": "search",
"time": "329.824",
"value": "SALISBURY EXETER"
},
{
"operation": "Now on",
"time": "331.334",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "339.063",
"value": "Southern Railway routes west of Salisbury"
},
{
"operation": "Now on",
"time": "340.161",
"value": "Southern Railway routes west of Salisbury"
},
{
"operation": "search",
"time": "374.696",
"value": "SOUTHERN RAILWAY ROUTES EAST"
},
{
"operation": "Now on",
"time": "376.753",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "408.003",
"value": "British Rail"
},
{
"operation": "Now on",
"time": "409.086",
"value": "British Rail"
},
{
"operation": "Page search",
"time": "433.674",
"value": "SALISBURY"
},
{
"operation": "hyperlink",
"time": "473.715",
"value": "Rail transport in Great Britain"
},
{
"operation": "Now on",
"time": "474.754",
"value": "Rail transport in Great Britain"
},
{
"operation": "Page search",
"time": "508.027",
"value": "SALISBURY"
},
{
"operation": "search",
"time": "548.813",
"value": "SALISBURY"
},
{
"operation": "Now on",
"time": "549.763",
"value": "Salisbury"
},
{
"operation": "hyperlink",
"time": "561.798",
"value": "10.3 Railways Salisbury_section_17"
},
{
"operation": "hyperlink",
"time": "591.916",
"value": "West of England Main Line"
},
{
"operation": "Now on",
"time": "592.698",
"value": "West of England line"
},
{
"operation": "Highlighting",
"time": "621.512",
"value": "West of England line_sentence_19"
},
{
"operation": "Highlighting",
"time": "621.514",
"value": "West of England line_title"
},
{
"operation": "Highlighting",
"time": "621.514",
"value": "West of England line_section_1"
},
{
"operation": "search",
"time": "650.352",
"value": "Contains...no major population centre until reaching Exeter"
},
{
"operation": "Now on",
"time": "652.491",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "657",
"value": "Southern Railway routes west of Salisbury"
},
{
"operation": "Now on",
"time": "657.333",
"value": "Southern Railway routes west of Salisbury"
},
{
"operation": "Page search",
"time": "663.316",
"value": "no major population centre until reaching Exeter"
},
{
"operation": "Highlighting",
"time": "671.925",
"value": "Southern Railway routes west of Salisbury_sentence_2"
},
{
"operation": "Highlighting",
"time": "675.762",
"value": "Southern Railway routes west of Salisbury_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "730.65",
"value": "Southern Railway routes west of Salisbury_sentence_3"
},
{
"operation": "finish",
"time": "737.874",
"value": "finish"
}
]
}
|
exfever_8754
|
Lana Wachowski and Lilly Wachowski achieved fame with their second film The Matrix, a major box office success for which they won the annual awards given by the American Academy of Science Fiction, Fantasy & Horror Films.
|
{
"evidence": [
"The Wachowskis",
"Saturn Award for Best Director"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nLana Wachowski and Lilly Wachowski achieved fame with their second film The Matrix, a major box office success for which they won the annual awards given by the American Academy of Science Fiction, Fantasy & Horror Films.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Wachowskis",
"Saturn Award for Best Director"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 589,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
33848
|
Fort Smith Museum of History is located at 320 Rogers Avenue in Fort Smith, Arkansas, and consists of a large four-story brick building with typical early-20th-century commercial styling, and in 1979 the museum was listed on the National Register of Historic Places, which is the official list of districts, sites, buildings, deemed worthy of preservation for their historical significance.
|
{
"evidence": [
"Fort Smith Museum of History_sentence_0",
"Fort Smith Museum of History_sentence_3",
"National Register of Historic Places_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:\nFort Smith Museum of History is located at 320 Rogers Avenue in Fort Smith, Arkansas, and consists of a large four-story brick building with typical early-20th-century commercial styling, and in 1979 the museum was listed on the National Register of Historic Places, which is the official list of districts, sites, buildings, deemed worthy of preservation for their historical significance.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Fort Smith Museum of History_sentence_0",
"Fort Smith Museum of History_sentence_3",
"National Register of Historic Places_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 590,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.723",
"value": "?search="
},
{
"operation": "search",
"time": "10.938",
"value": "Fort Smith Museum of History"
},
{
"operation": "Now on",
"time": "11.52",
"value": "Fort Smith Museum of History"
},
{
"operation": "Highlighting",
"time": "16.69",
"value": "Fort Smith Museum of History_sentence_0"
},
{
"operation": "Highlighting",
"time": "37.056",
"value": "Fort Smith Museum of History_sentence_3"
},
{
"operation": "hyperlink",
"time": "49.334",
"value": "National Register of Historic Places"
},
{
"operation": "Now on",
"time": "50.103",
"value": "National Register of Historic Places"
},
{
"operation": "Highlighting",
"time": "111.122",
"value": "National Register of Historic Places_sentence_2"
},
{
"operation": "finish",
"time": "136.399",
"value": "finish"
}
]
}
|
exfever_8714
|
Outlander is a historical drama television series premiered on August 9, 2014 starring Caitríona Balfe as a fictional character in the Outlander, who is a married World War II nurse visiting Scotland who finds herself transported from 1945 back to 1730.
|
{
"evidence": [
"Outlander (TV series)",
"Claire Fraser (character)"
],
"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:\nOutlander is a historical drama television series premiered on August 9, 2014 starring Caitríona Balfe as a fictional character in the Outlander, who is a married World War II nurse visiting Scotland who finds herself transported from 1945 back to 1730.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Outlander (TV series)",
"Claire Fraser (character)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 591,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
57459
|
Horticultural societies typically engage in small-scale gardening with simple tools, these methods allow for higher population densities while agriculture is the herding and breeding of domestic animals.
|
{
"evidence": [
"Subsistence pattern_sentence_21"
],
"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:\nHorticultural societies typically engage in small-scale gardening with simple tools, these methods allow for higher population densities while agriculture is the herding and breeding of domestic animals.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Subsistence pattern_sentence_21"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 592,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.749",
"value": "undefined"
},
{
"operation": "search",
"time": "5.052",
"value": "Subsistence pattern"
},
{
"operation": "Now on",
"time": "5.714",
"value": "Subsistence pattern"
},
{
"operation": "Highlighting",
"time": "15.723",
"value": "Subsistence pattern_sentence_21"
},
{
"operation": "finish",
"time": "26.986",
"value": "finish"
}
]
}
|
exfever_2801
|
Generation X is the most influential generations according to studies.
|
{
"evidence": [
"Generation X",
"Millennials"
],
"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:\nGeneration X is the most influential generations according to studies.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Generation X",
"Millennials"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 593,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7092
|
Susan Sarandon is an American activist who stars in an American drama film written by Norman Wexler and directed by John G. Avildsen, starring an American actor known as a character actor on Everybody Loves Raymond.
|
{
"evidence": [
"Susan Sarandon",
"Joe (1970 film)",
"Peter Boyle"
],
"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:\nSusan Sarandon is an American activist who stars in an American drama film written by Norman Wexler and directed by John G. Avildsen, starring an American actor known as a character actor on Everybody Loves Raymond.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Susan Sarandon",
"Joe (1970 film)",
"Peter Boyle"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 594,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4028
|
Christopher Lloyd is known for portraying Jim Ignatowski in the comedy series that originally aired on CBC.
|
{
"evidence": [
"Christopher Lloyd",
"Taxi (TV series)"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nChristopher Lloyd is known for portraying Jim Ignatowski in the comedy series that originally aired on CBC.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Christopher Lloyd",
"Taxi (TV series)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 595,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6352
|
Django Unchained is a 2012 revisionist Western film starring an American actor who has appeared in several television series, including Justified.
|
{
"evidence": [
"Django Unchained",
"Walton Goggins"
],
"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:\nDjango Unchained is a 2012 revisionist Western film starring an American actor who has appeared in several television series, including Justified.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Django Unchained",
"Walton Goggins"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 596,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
42806
|
Cornelis de Witt is an Irish regent, governor, and landlord who is also known as Cornelis de Witt and was the son of Jacob de Witt, a burgomaster of Dordrecht.
|
{
"evidence": [
"Cornelis de Witt_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:\nCornelis de Witt is an Irish regent, governor, and landlord who is also known as Cornelis de Witt and was the son of Jacob de Witt, a burgomaster of Dordrecht.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Cornelis de Witt_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 597,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.688",
"value": "?search="
},
{
"operation": "search",
"time": "49.001",
"value": "Cornelis de Witt"
},
{
"operation": "Now on",
"time": "49.616",
"value": "Cornelis de Witt"
},
{
"operation": "Highlighting",
"time": "72.108",
"value": "Cornelis de Witt_sentence_0"
},
{
"operation": "finish",
"time": "86.8",
"value": "finish"
}
]
}
|
exfever_6878
|
AC/DC is an Australian jazz band whose music has been variously described as a loosely defined subgenre of rock music typified by heavy use of aggressive vocals, distorted electric guitars, bass guitar, and drums, sometimes accompanied by keyboards.
|
{
"evidence": [
"AC/DC",
"Hard rock"
],
"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:\nAC/DC is an Australian jazz band whose music has been variously described as a loosely defined subgenre of rock music typified by heavy use of aggressive vocals, distorted electric guitars, bass guitar, and drums, sometimes accompanied by keyboards.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"AC/DC",
"Hard rock"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 598,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_978
|
Bridge of Spies is a 2015 historical drama film directed and co-produced by Steven Spielberg, starring an American actor who made his breakthrough roles in a series of comedy films including a 1984 American sex comedy film directed by Neal Israel.
|
{
"evidence": [
"Bridge of Spies (film)",
"Tom Hanks",
"Bachelor Party (1984 film)"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBridge of Spies is a 2015 historical drama film directed and co-produced by Steven Spielberg, starring an American actor who made his breakthrough roles in a series of comedy films including a 1984 American sex comedy film directed by Neal Israel.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Bridge of Spies (film)",
"Tom Hanks",
"Bachelor Party (1984 film)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 599,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
75244
|
Bojan Jamina (born February 5, 1979) is a retired midfielder who played for FK Slavija Sarajevo and Celik Zenica.
|
{
"evidence": [
"Bojan Jamina_sentence_0",
"Bojan Jamina_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:\nBojan Jamina (born February 5, 1979) is a retired midfielder who played for FK Slavija Sarajevo and Celik Zenica.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Bojan Jamina_sentence_0",
"Bojan Jamina_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 600,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.108",
"value": "?search="
},
{
"operation": "search",
"time": "7.546",
"value": "Bojan Jamina"
},
{
"operation": "Now on",
"time": "9.387",
"value": "Bojan Jamina"
},
{
"operation": "Highlighting",
"time": "11.343",
"value": "Bojan Jamina_sentence_0"
},
{
"operation": "Highlighting",
"time": "25.565",
"value": "Bojan Jamina_sentence_2"
},
{
"operation": "finish",
"time": "703.068",
"value": "finish"
}
]
}
|
71923
|
In 1894, the first race was organised between Lévis and the Port of Quebec (the oldest port in Canada) at the first Quebec Winter Carnival; while the Montreal races were held during the Fête des Neiges de Montréal from 1988 to 1992.
|
{
"evidence": [
"Ice canoeing_sentence_7",
"Port of Quebec_sentence_1",
"Ice canoeing_sentence_8"
],
"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 1894, the first race was organised between Lévis and the Port of Quebec (the oldest port in Canada) at the first Quebec Winter Carnival; while the Montreal races were held during the Fête des Neiges de Montréal from 1988 to 1992.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Ice canoeing_sentence_7",
"Port of Quebec_sentence_1",
"Ice canoeing_sentence_8"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 601,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.695",
"value": "undefined"
},
{
"operation": "search",
"time": "38.536",
"value": "Contains...Quebec Winter Carnival"
},
{
"operation": "Now on",
"time": "40.275",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "52.464",
"value": "Ice canoeing"
},
{
"operation": "Now on",
"time": "53.78",
"value": "Ice canoeing"
},
{
"operation": "Highlighting",
"time": "79.268",
"value": "Ice canoeing_sentence_7"
},
{
"operation": "hyperlink",
"time": "82.972",
"value": "Port of Quebec"
},
{
"operation": "Now on",
"time": "83.492",
"value": "Port of Quebec"
},
{
"operation": "Highlighting",
"time": "86.434",
"value": "Port of Quebec_sentence_1"
},
{
"operation": "back-button-clicked",
"time": "88.695",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "88.782",
"value": "Ice canoeing"
},
{
"operation": "Highlighting",
"time": "94.144",
"value": "Ice canoeing_sentence_8"
},
{
"operation": "finish",
"time": "115.387",
"value": "finish"
}
]
}
|
73487
|
New South Wales Team won against South Canterbury on August 23, 1923 with a score of 23-10.
|
{
"evidence": [
"New South Wales cricket team_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:\nNew South Wales Team won against South Canterbury on August 23, 1923 with a score of 23-10.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"New South Wales cricket team_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 602,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.659",
"value": "undefined"
},
{
"operation": "search",
"time": "19.889",
"value": "Contains...1923 New South Wales Team"
},
{
"operation": "Now on",
"time": "22.313",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "32.6",
"value": "New South Wales cricket team"
},
{
"operation": "Now on",
"time": "34.025",
"value": "New South Wales cricket team"
},
{
"operation": "Highlighting",
"time": "40.496",
"value": "New South Wales cricket team_sentence_0"
},
{
"operation": "search",
"time": "158.826",
"value": "New South Wales cricket team"
},
{
"operation": "Now on",
"time": "159.995",
"value": "New South Wales cricket team"
},
{
"operation": "Page search",
"time": "212.81",
"value": "1923"
},
{
"operation": "search",
"time": "216.451",
"value": "New South Wales Cricket Team"
},
{
"operation": "Now on",
"time": "218.576",
"value": "New South Wales cricket team"
},
{
"operation": "hyperlink",
"time": "249.36",
"value": "Sheffield Shield"
},
{
"operation": "Now on",
"time": "250.775",
"value": "Sheffield Shield"
},
{
"operation": "search",
"time": "279.96",
"value": "1923 Sheffield Shield"
},
{
"operation": "Now on",
"time": "281.728",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "284.974",
"value": "1923–24 Sheffield Shield season"
},
{
"operation": "Now on",
"time": "285.695",
"value": "1923–24 Sheffield Shield season"
},
{
"operation": "hyperlink",
"time": "301.98",
"value": "New South Wales"
},
{
"operation": "Now on",
"time": "303.027",
"value": "New South Wales cricket team"
},
{
"operation": "finish",
"time": "329.816",
"value": "finish"
}
]
}
|
exfever_4455
|
Dennis Hopper has appeared in a 1956 American epic Western drama film from a screenplay adapted by an American film director and screenwriter.
|
{
"evidence": [
"Dennis Hopper",
"Giant (1956 film)",
"Fred Guiol"
],
"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:\nDennis Hopper has appeared in a 1956 American epic Western drama film from a screenplay adapted by an American film director and screenwriter.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Dennis Hopper",
"Giant (1956 film)",
"Fred Guiol"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 603,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2680
|
Stranger than Fiction is a 2006 film directed by Marc Forster and starring a European actor who established himself in the mid-1990s as a cast member on the NBC sketch comedy show Saturday Night Live.
|
{
"evidence": [
"Stranger than Fiction (2006 film)",
"Will Ferrell"
],
"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:\nStranger than Fiction is a 2006 film directed by Marc Forster and starring a European actor who established himself in the mid-1990s as a cast member on the NBC sketch comedy show Saturday Night Live.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Stranger than Fiction (2006 film)",
"Will Ferrell"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 604,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_96
|
The Connaught Tunnel is in southeastern British Columbia, on the segment between Revelstoke and a small sawmilling community located on Highway 1.
|
{
"evidence": [
"Connaught Tunnel",
"Donald, British Columbia"
],
"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 Connaught Tunnel is in southeastern British Columbia, on the segment between Revelstoke and a small sawmilling community located on Highway 1.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Connaught Tunnel",
"Donald, British Columbia"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 605,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
72542
|
Costa Rican customs is located at the west end of a former railroad bridge which crosses the Rio Panamanian (a river that flows from the Cordillera Talamanca to the Caribbean Sea northeast of Sixaola) at the border, where, when crossing, tourists must clear both Costa Rican and Panamanian customs.
|
{
"evidence": [
"List of rivers of Panama_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:\nCosta Rican customs is located at the west end of a former railroad bridge which crosses the Rio Panamanian (a river that flows from the Cordillera Talamanca to the Caribbean Sea northeast of Sixaola) at the border, where, when crossing, tourists must clear both Costa Rican and Panamanian customs.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"List of rivers of Panama_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 606,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.844",
"value": "undefined"
},
{
"operation": "search",
"time": "72.063",
"value": "west end of a former railroad bridge"
},
{
"operation": "Now on",
"time": "75.681",
"value": "undefined"
},
{
"operation": "search",
"time": "114.178",
"value": "panama river"
},
{
"operation": "Now on",
"time": "115.853",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "122.551",
"value": "Grande River (Panama)"
},
{
"operation": "Now on",
"time": "123.834",
"value": "Grande River (Panama)"
},
{
"operation": "hyperlink",
"time": "132.336",
"value": "List of rivers of Panama"
},
{
"operation": "Now on",
"time": "133.544",
"value": "List of rivers of Panama"
},
{
"operation": "Highlighting",
"time": "158.531",
"value": "List of rivers of Panama_sentence_0"
},
{
"operation": "finish",
"time": "164.649",
"value": "finish"
}
]
}
|
26412
|
Billie Carleton died of an apparent overdose of heroin after attending the Victory Ball at the Royal Albert Hall in 1920.
|
{
"evidence": [
"Billie Carleton_sentence_18"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBillie Carleton died of an apparent overdose of heroin after attending the Victory Ball at the Royal Albert Hall in 1920.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Billie Carleton_sentence_18"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 607,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.642",
"value": "?search="
},
{
"operation": "search",
"time": "19.583",
"value": "Billie Carleton"
},
{
"operation": "Now on",
"time": "22.112",
"value": "Billie Carleton"
},
{
"operation": "Highlighting",
"time": "73.626",
"value": "Billie Carleton_sentence_18"
},
{
"operation": "finish",
"time": "103.008",
"value": "finish"
}
]
}
|
exfever_3008
|
Heather Joan Graham’s second starring role in a 1988 American teen comedy film written by Neil Tolkin.
|
{
"evidence": [
"Heather Graham",
"License to Drive"
],
"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:\nHeather Joan Graham’s second starring role in a 1988 American teen comedy film written by Neil Tolkin.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Heather Graham",
"License to Drive"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 608,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3863
|
The Bostonians is a 1985 romantic drama period film directed by James Ivory and starring Vanessa Redgrave, Christopher Reeve, Jessica Tandy, and the American actress who has appeared in over twenty films and television shows, including four James Ivory projects.
|
{
"evidence": [
"The Bostonians (film)",
"Madeleine Potter"
],
"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 Bostonians is a 1985 romantic drama period film directed by James Ivory and starring Vanessa Redgrave, Christopher Reeve, Jessica Tandy, and the American actress who has appeared in over twenty films and television shows, including four James Ivory projects.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Bostonians (film)",
"Madeleine Potter"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 609,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
88785
|
Dénes Kemény (born 14 June 1954 in Budapest) and helped the Hungarian team won at least a medal in 24 of its 29 major tournaments and was granted Hungarian Sport President of the Year award five times
|
{
"evidence": [
"Dénes Kemény_sentence_2",
"Dénes Kemény_sentence_3",
"Dénes Kemény_sentence_8"
],
"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:\nDénes Kemény (born 14 June 1954 in Budapest) and helped the Hungarian team won at least a medal in 24 of its 29 major tournaments and was granted Hungarian Sport President of the Year award five times\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Dénes Kemény_sentence_2",
"Dénes Kemény_sentence_3",
"Dénes Kemény_sentence_8"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 610,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.521",
"value": "?search="
},
{
"operation": "Now on",
"time": "4.486",
"value": "?search="
},
{
"operation": "search",
"time": "19.752",
"value": "Dénes Kemény"
},
{
"operation": "Now on",
"time": "21.007",
"value": "Dénes Kemény"
},
{
"operation": "Highlighting",
"time": "93.003",
"value": "Dénes Kemény_sentence_2"
},
{
"operation": "Highlighting",
"time": "101.304",
"value": "Dénes Kemény_sentence_3"
},
{
"operation": "Highlighting",
"time": "118.897",
"value": "Dénes Kemény_sentence_8"
},
{
"operation": "finish",
"time": "124.906",
"value": "finish"
}
]
}
|
exfever_6724
|
The Great Buck Howard is an American comedy-drama film starring an actor who is one of the least popular and unrecognizable film stars worldwide.
|
{
"evidence": [
"The Great Buck Howard",
"Tom Hanks"
],
"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 Great Buck Howard is an American comedy-drama film starring an actor who is one of the least popular and unrecognizable film stars worldwide.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Great Buck Howard",
"Tom Hanks"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 611,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
12342
|
Maryvale Station is where the unusual rock formation Chambers Pillar is situated and the property has an average stocking rate of approximately 7,000 head of cattle.
|
{
"evidence": [
"Maryvale Station_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:\nMaryvale Station is where the unusual rock formation Chambers Pillar is situated and the property has an average stocking rate of approximately 7,000 head of cattle.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Maryvale Station_sentence_7"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 612,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.682",
"value": "?search="
},
{
"operation": "search",
"time": "15.395",
"value": "Maryvale Station"
},
{
"operation": "Now on",
"time": "16.233",
"value": "Maryvale Station"
},
{
"operation": "Highlighting",
"time": "24.493",
"value": "Maryvale Station_sentence_6"
},
{
"operation": "Highlighting",
"time": "28.678",
"value": "Maryvale Station_sentence_7"
},
{
"operation": "Highlighting deleted",
"time": "32.396",
"value": "Maryvale Station_sentence_6"
},
{
"operation": "finish",
"time": "47.505",
"value": "finish"
}
]
}
|
exfever_8865
|
The Fate of the Furious is an Filipino action film starring an American actor who also appear in a supporting role in an American epic war film directed by Steven Spielberg.
|
{
"evidence": [
"The Fate of the Furious",
"Vin Diesel",
"Saving Private Ryan"
],
"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 an Filipino action film starring an American actor who also appear in a supporting role in an American epic war film directed by Steven Spielberg.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Fate of the Furious",
"Vin Diesel",
"Saving Private Ryan"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 613,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
35405
|
The center part of the Dartmoor (one of the places where exposed masses of the Cornubian batholith granite rock can be seen) outcrop is a medium-grained granite with many very large megacrysts, while the main outcrop consist of coarse-grained megacrystic granite with small megacrysts.
|
{
"evidence": [
"Dartmoor_sentence_4",
"Cornubian batholith_sentence_1",
"Dartmoor_sentence_2",
"Dartmoor_sentence_5",
"Dartmoor_sentence_12",
"Geology of Dartmoor National Park_sentence_2",
"Tor (rock formation)_sentence_7",
"Tor (rock formation)_sentence_16",
"Cornubian batholith_sentence_6"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe center part of the Dartmoor (one of the places where exposed masses of the Cornubian batholith granite rock can be seen) outcrop is a medium-grained granite with many very large megacrysts, while the main outcrop consist of coarse-grained megacrystic granite with small megacrysts.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Dartmoor_sentence_4",
"Cornubian batholith_sentence_1",
"Dartmoor_sentence_2",
"Dartmoor_sentence_5",
"Dartmoor_sentence_12",
"Geology of Dartmoor National Park_sentence_2",
"Tor (rock formation)_sentence_7",
"Tor (rock formation)_sentence_16",
"Cornubian batholith_sentence_6"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 614,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.906",
"value": "?search="
},
{
"operation": "search",
"time": "21.989",
"value": "dartmoor"
},
{
"operation": "Now on",
"time": "26.01",
"value": "Dartmoor"
},
{
"operation": "Page search",
"time": "65.589",
"value": "megacryst"
},
{
"operation": "hyperlink",
"time": "90.701",
"value": "Dartmoor (disambiguation)"
},
{
"operation": "Now on",
"time": "91.573",
"value": "Dartmoor (disambiguation)"
},
{
"operation": "hyperlink",
"time": "110.59",
"value": "Dartmoor"
},
{
"operation": "Now on",
"time": "111.475",
"value": "Dartmoor"
},
{
"operation": "Highlighting",
"time": "119.936",
"value": "Dartmoor_sentence_4"
},
{
"operation": "Page search",
"time": "139.671",
"value": "batho"
},
{
"operation": "search",
"time": "187.974",
"value": "Cornubian batholith"
},
{
"operation": "Now on",
"time": "190.549",
"value": "Cornubian batholith"
},
{
"operation": "Highlighting",
"time": "195.645",
"value": "Cornubian batholith_sentence_1"
},
{
"operation": "hyperlink",
"time": "209.388",
"value": "Dartmoor"
},
{
"operation": "Now on",
"time": "210.517",
"value": "Dartmoor"
},
{
"operation": "hyperlink",
"time": "230.505",
"value": "Geology of Dartmoor National Park"
},
{
"operation": "Now on",
"time": "231.92",
"value": "Geology of Dartmoor National Park"
},
{
"operation": "Page search",
"time": "249.539",
"value": "megacryst"
},
{
"operation": "Page search",
"time": "267.131",
"value": "outcrop"
},
{
"operation": "Page search",
"time": "301.159",
"value": "center"
},
{
"operation": "Page search",
"time": "308.604",
"value": "cent"
},
{
"operation": "Highlighting",
"time": "321.159",
"value": "Geology of Dartmoor National Park_sentence_37"
},
{
"operation": "Highlighting",
"time": "321.164",
"value": "Geology of Dartmoor National Park_section_7"
},
{
"operation": "Highlighting deleted",
"time": "323.769",
"value": "Geology of Dartmoor National Park_sentence_37"
},
{
"operation": "Highlighting deleted",
"time": "323.776",
"value": "Geology of Dartmoor National Park_section_7"
},
{
"operation": "Page search",
"time": "357.741",
"value": "medium"
},
{
"operation": "hyperlink",
"time": "373.727",
"value": "Dartmoor"
},
{
"operation": "Now on",
"time": "375.128",
"value": "Dartmoor"
},
{
"operation": "Page search",
"time": "378.726",
"value": "medium"
},
{
"operation": "Highlighting",
"time": "399.877",
"value": "Dartmoor_sentence_2"
},
{
"operation": "Highlighting",
"time": "407.824",
"value": "Dartmoor_sentence_5"
},
{
"operation": "Highlighting",
"time": "425.835",
"value": "Dartmoor_sentence_12"
},
{
"operation": "hyperlink",
"time": "434.2",
"value": "Geology of Dartmoor National Park"
},
{
"operation": "Now on",
"time": "435.04",
"value": "Geology of Dartmoor National Park"
},
{
"operation": "Highlighting",
"time": "439.38",
"value": "Geology of Dartmoor National Park_sentence_2"
},
{
"operation": "Page search",
"time": "449.494",
"value": "mega"
},
{
"operation": "hyperlink",
"time": "470.716",
"value": "Dartmoor"
},
{
"operation": "Now on",
"time": "471.821",
"value": "Dartmoor"
},
{
"operation": "hyperlink",
"time": "478.153",
"value": "moorland"
},
{
"operation": "Now on",
"time": "479.645",
"value": "Moorland"
},
{
"operation": "hyperlink",
"time": "486.314",
"value": "Dartmoor"
},
{
"operation": "Now on",
"time": "487.047",
"value": "Dartmoor"
},
{
"operation": "search",
"time": "505.186",
"value": "dartmoor outcrop"
},
{
"operation": "Now on",
"time": "507.106",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "512.201",
"value": "Tors of Dartmoor"
},
{
"operation": "Now on",
"time": "513.263",
"value": "Tor (rock formation)"
},
{
"operation": "Highlighting",
"time": "523.661",
"value": "Tor (rock formation)_sentence_7"
},
{
"operation": "Highlighting",
"time": "530.792",
"value": "Tor (rock formation)_sentence_16"
},
{
"operation": "search",
"time": "546.495",
"value": "Cornubian batholith"
},
{
"operation": "Now on",
"time": "547.375",
"value": "Cornubian batholith"
},
{
"operation": "Highlighting",
"time": "561.797",
"value": "Cornubian batholith_sentence_6"
},
{
"operation": "finish",
"time": "593.077",
"value": "finish"
}
]
}
|
exfever_2679
|
Mediterranean cuisine is the food and methods of preparation used by the people of the region of lands around the Mediterranean Sea that have mostly a dry summer climate characterized by dry summers and mild, wet winters.
|
{
"evidence": [
"Mediterranean cuisine",
"Mediterranean Basin",
"Mediterranean climate"
],
"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:\nMediterranean cuisine is the food and methods of preparation used by the people of the region of lands around the Mediterranean Sea that have mostly a dry summer climate characterized by dry summers and mild, wet winters.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Mediterranean cuisine",
"Mediterranean Basin",
"Mediterranean climate"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 615,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4264
|
Laura Linney is an American actress who received nominations for a 1953 play by American playwright Mel Gibson.
|
{
"evidence": [
"Laura Linney",
"The Crucible"
],
"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:\nLaura Linney is an American actress who received nominations for a 1953 play by American playwright Mel Gibson.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Laura Linney",
"The Crucible"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 616,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
10464
|
Guandu metro stationlocated in Taiwan was originally opened on 25 December 1901, discontinued in 1988 and the five millionth EasyCard was sold at the station in 2005.
|
{
"evidence": [
"Guandu metro station_sentence_0",
"Guandu metro station_sentence_6",
"Guandu metro station_sentence_8",
"Guandu metro station_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:\nGuandu metro stationlocated in Taiwan was originally opened on 25 December 1901, discontinued in 1988 and the five millionth EasyCard was sold at the station in 2005.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Guandu metro station_sentence_0",
"Guandu metro station_sentence_6",
"Guandu metro station_sentence_8",
"Guandu metro station_sentence_9"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 617,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.698",
"value": "?search="
},
{
"operation": "search",
"time": "15.57",
"value": "Guandu metro station"
},
{
"operation": "Now on",
"time": "16.111",
"value": "Guandu metro station"
},
{
"operation": "Highlighting",
"time": "21.193",
"value": "Guandu metro station_sentence_0"
},
{
"operation": "Highlighting",
"time": "30.536",
"value": "Guandu metro station_sentence_6"
},
{
"operation": "Highlighting",
"time": "40.16",
"value": "Guandu metro station_sentence_8"
},
{
"operation": "Highlighting",
"time": "44.712",
"value": "Guandu metro station_sentence_9"
},
{
"operation": "finish",
"time": "76.263",
"value": "finish"
}
]
}
|
32656
|
WNDZ (750 AM) is a radio station broadcasting a brokered programming or "time-buy" and "blocktime" format licensed to Portage, Indiana, United States and serving the Chicago area.
|
{
"evidence": [
"WNDZ_sentence_0",
"Brokered programming_sentence_0",
"WNDZ_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:\nWNDZ (750 AM) is a radio station broadcasting a brokered programming or \"time-buy\" and \"blocktime\" format licensed to Portage, Indiana, United States and serving the Chicago area.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"WNDZ_sentence_0",
"Brokered programming_sentence_0",
"WNDZ_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 618,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.755",
"value": "?search="
},
{
"operation": "search",
"time": "11.64",
"value": "WNDZ"
},
{
"operation": "Now on",
"time": "12.513",
"value": "WNDZ"
},
{
"operation": "Highlighting",
"time": "20.904",
"value": "WNDZ_sentence_0"
},
{
"operation": "hyperlink",
"time": "30.417",
"value": "brokered programming"
},
{
"operation": "Now on",
"time": "31.137",
"value": "Brokered programming"
},
{
"operation": "Highlighting",
"time": "32.731",
"value": "Brokered programming_sentence_0"
},
{
"operation": "back-button-clicked",
"time": "37.824",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "37.922",
"value": "WNDZ"
},
{
"operation": "Highlighting",
"time": "43.657",
"value": "WNDZ_sentence_1"
},
{
"operation": "finish",
"time": "72.92",
"value": "finish"
}
]
}
|
16623
|
Koopmans, born in Amsterdam, organised his first Alternative Elfstedentocht in 1975 in Finland; then later, the Alternative Elfstedentochten were organised in various countries.
|
{
"evidence": [
"Aart Koopmans_sentence_2",
"Aart Koopmans_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:\nKoopmans, born in Amsterdam, organised his first Alternative Elfstedentocht in 1975 in Finland; then later, the Alternative Elfstedentochten were organised in various countries.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Aart Koopmans_sentence_2",
"Aart Koopmans_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 619,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.117",
"value": "?search="
},
{
"operation": "search",
"time": "17.473",
"value": "Koopmans"
},
{
"operation": "Now on",
"time": "19.205",
"value": "Koopmans"
},
{
"operation": "hyperlink",
"time": "25.897",
"value": "Koopman"
},
{
"operation": "Now on",
"time": "27.305",
"value": "Koopman"
},
{
"operation": "Page search",
"time": "37.17",
"value": "amsterdam"
},
{
"operation": "search",
"time": "46.693",
"value": "Alternative Elfstedentocht"
},
{
"operation": "Now on",
"time": "48.211",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "54.309",
"value": "Alternative Elfstedentocht Elfstedentocht_section_7"
},
{
"operation": "Now on",
"time": "55.431",
"value": "Elfstedentocht#Alternative Elfstedentocht Elfstedentocht section 7"
},
{
"operation": "Page search",
"time": "64.655",
"value": "koopman"
},
{
"operation": "search",
"time": "87.765",
"value": "Alternative Elfstedentocht in 1975"
},
{
"operation": "Now on",
"time": "88.715",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "91.557",
"value": "Aart Koopmans"
},
{
"operation": "Now on",
"time": "92.163",
"value": "Aart Koopmans"
},
{
"operation": "Page search",
"time": "98.426",
"value": "Alternative Elfstedentocht in 1975"
},
{
"operation": "Highlighting",
"time": "103.518",
"value": "Aart Koopmans_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "112.542",
"value": "Aart Koopmans_sentence_0"
},
{
"operation": "Highlighting",
"time": "113.351",
"value": "Aart Koopmans_sentence_2"
},
{
"operation": "Highlighting",
"time": "130.542",
"value": "Aart Koopmans_sentence_3"
},
{
"operation": "search",
"time": "159.025",
"value": "koopmans"
},
{
"operation": "Now on",
"time": "160.533",
"value": "Koopmans"
},
{
"operation": "finish",
"time": "167.984",
"value": "finish"
}
]
}
|
46381
|
Tomorrow's Harvest peaked on seventh in the UK charts.
|
{
"evidence": [
"Tomorrow's Harvest_sentence_43"
],
"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:\nTomorrow's Harvest peaked on seventh in the UK charts.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Tomorrow's Harvest_sentence_43"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 620,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.355",
"value": "?search="
},
{
"operation": "search",
"time": "7.102",
"value": "Tomorrow's Harvest"
},
{
"operation": "Now on",
"time": "8.484",
"value": "Tomorrow's Harvest"
},
{
"operation": "Page search",
"time": "23.689",
"value": "peak"
},
{
"operation": "Page search",
"time": "111.542",
"value": "UK"
},
{
"operation": "Highlighting",
"time": "201.156",
"value": "Tomorrow's Harvest_sentence_43"
},
{
"operation": "finish",
"time": "242.02",
"value": "finish"
}
]
}
|
exfever_565
|
Match Point is a 2005 film written and directed by Woody Allen, whose genre combines the plot with psychological narratives or thrilling settings.
|
{
"evidence": [
"Match Point",
"Psychological thriller"
],
"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:\nMatch Point is a 2005 film written and directed by Woody Allen, whose genre combines the plot with psychological narratives or thrilling settings.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Match Point",
"Psychological thriller"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 621,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8005
|
Sheri Moon earned an acting award for her role in the House of 1000 Corpses.
|
{
"evidence": [
"House of 1000 Corpses",
"Bill Moseley"
],
"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:\nSheri Moon earned an acting award for her role in the House of 1000 Corpses.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"House of 1000 Corpses",
"Bill Moseley"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 622,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3220
|
Tom Hiddleston is an English actor who gained international fame portraying Loki in the Marvel Cinematic Universe, appearing in a 2013 American superhero film based on the character Thor from the brand name and primary imprint of Marvel Worldwide Inc.
|
{
"evidence": [
"Tom Hiddleston",
"Thor: The Dark World",
"Marvel Comics"
],
"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:\nTom Hiddleston is an English actor who gained international fame portraying Loki in the Marvel Cinematic Universe, appearing in a 2013 American superhero film based on the character Thor from the brand name and primary imprint of Marvel Worldwide Inc.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Tom Hiddleston",
"Thor: The Dark World",
"Marvel Comics"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 623,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
24925
|
BOSS GP moved to Europe (a continent located entirely in the Northern Hemisphere)and saw girds of around 12-15 cars, people split off to become BOSS GP.
|
{
"evidence": [
"BOSS GP_sentence_4",
"BOSS GP_sentence_6",
"BOSS GP_sentence_8",
"BOSS GP_sentence_7",
"Europe_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:\nBOSS GP moved to Europe (a continent located entirely in the Northern Hemisphere)and saw girds of around 12-15 cars, people split off to become BOSS GP.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"BOSS GP_sentence_4",
"BOSS GP_sentence_6",
"BOSS GP_sentence_8",
"BOSS GP_sentence_7",
"Europe_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 624,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.419",
"value": "undefined"
},
{
"operation": "search",
"time": "9.014",
"value": "BOSS GP"
},
{
"operation": "Now on",
"time": "11.413",
"value": "BOSS GP"
},
{
"operation": "Highlighting",
"time": "18.635",
"value": "BOSS GP_sentence_4"
},
{
"operation": "Highlighting",
"time": "28.398",
"value": "BOSS GP_sentence_6"
},
{
"operation": "Highlighting",
"time": "41.927",
"value": "BOSS GP_sentence_8"
},
{
"operation": "Highlighting",
"time": "52.041",
"value": "BOSS GP_sentence_7"
},
{
"operation": "search",
"time": "63.28",
"value": "Europe"
},
{
"operation": "Now on",
"time": "67.065",
"value": "Europe"
},
{
"operation": "Highlighting",
"time": "69.32",
"value": "Europe_sentence_2"
},
{
"operation": "finish",
"time": "101.465",
"value": "finish"
}
]
}
|
67202
|
Singer, songwriter and rapper Coco Jones released four singles in 2009 that made it to the top ten charts.
|
{
"evidence": [
"Coco Jones discography_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:\nSinger, songwriter and rapper Coco Jones released four singles in 2009 that made it to the top ten charts.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Coco Jones discography_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 625,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.134",
"value": "?search="
},
{
"operation": "search",
"time": "13.793",
"value": "Coco Jones"
},
{
"operation": "Now on",
"time": "16.598",
"value": "Coco Jones"
},
{
"operation": "hyperlink",
"time": "157.75",
"value": "Coco Jones discography"
},
{
"operation": "Now on",
"time": "158.386",
"value": "Coco Jones discography"
},
{
"operation": "Highlighting",
"time": "259.277",
"value": "Coco Jones discography_sentence_0"
},
{
"operation": "finish",
"time": "262.356",
"value": "finish"
}
]
}
|
7980
|
The Ballynahatty woman is the name given to a prehistoric female human found in the Ballynahatty townland near Belfast in 1855 who is estimated to have lived about 5,000 years ago.
|
{
"evidence": [
"Ballynahatty woman_sentence_0",
"Ballynahatty woman_sentence_1"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Ballynahatty woman is the name given to a prehistoric female human found in the Ballynahatty townland near Belfast in 1855 who is estimated to have lived about 5,000 years ago.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Ballynahatty woman_sentence_0",
"Ballynahatty woman_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 626,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.183",
"value": "?search="
},
{
"operation": "search",
"time": "22.37",
"value": "Ballynahatty woman"
},
{
"operation": "Now on",
"time": "23.651",
"value": "Ballynahatty woman"
},
{
"operation": "Highlighting",
"time": "31.841",
"value": "Ballynahatty woman_sentence_0"
},
{
"operation": "Highlighting",
"time": "33.231",
"value": "Ballynahatty woman_sentence_1"
},
{
"operation": "finish",
"time": "45.907",
"value": "finish"
}
]
}
|
exfever_8805
|
Grunge was the most famous genre that emerged during the mid-1980s.
|
{
"evidence": [
"Grunge",
"Washington (state)",
"Western United States"
],
"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:\nGrunge was the most famous genre that emerged during the mid-1980s.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Grunge",
"Washington (state)",
"Western United States"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 627,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
26306
|
Karen, an independent artist who currently works full-time as the receptionist at a cemetery, wanted to apply for art college in 2006, but missed the deadline.
|
{
"evidence": [
"Planet Karen_sentence_22"
],
"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:\nKaren, an independent artist who currently works full-time as the receptionist at a cemetery, wanted to apply for art college in 2006, but missed the deadline.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Planet Karen_sentence_22"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 628,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.281",
"value": "?search="
},
{
"operation": "search",
"time": "16.622",
"value": "Contains...\"an independent artist who currently works full-time as the receptionist at a cemetery\""
},
{
"operation": "Now on",
"time": "18.13",
"value": "undefined"
},
{
"operation": "search",
"time": "47.754",
"value": "\"receptionist at a cemetery\""
},
{
"operation": "Now on",
"time": "49.495",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "51.422",
"value": "Planet Karen"
},
{
"operation": "Now on",
"time": "54.357",
"value": "Planet Karen"
},
{
"operation": "Page search",
"time": "57.349",
"value": "an independent artist who currently works full-time as the receptionist at a cemetery"
},
{
"operation": "Page search",
"time": "71.558",
"value": "an independent artist who currently works full-time as the receptionist at a cemetery"
},
{
"operation": "Highlighting",
"time": "95.952",
"value": "Planet Karen_sentence_22"
},
{
"operation": "finish",
"time": "124.431",
"value": "finish"
}
]
}
|
94233
|
Poropuntius consternans is a species of plant.
|
{
"evidence": [
"Poropuntius consternans_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:\nPoropuntius consternans is a species of plant.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Poropuntius consternans_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 629,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.71",
"value": "?search="
},
{
"operation": "search",
"time": "9.945",
"value": "Poropuntius consternans"
},
{
"operation": "Now on",
"time": "11.758",
"value": "Poropuntius consternans"
},
{
"operation": "Highlighting",
"time": "17.055",
"value": "Poropuntius consternans_sentence_0"
},
{
"operation": "finish",
"time": "43.894",
"value": "finish"
}
]
}
|
exfever_8379
|
The Bridges of Madison County was produced and directed by an American actor who rose to international fame with his role in the Dollars Trilogy, directed by an Italian film director credited as the creator of the Spaghetti Western genre.
|
{
"evidence": [
"The Bridges of Madison County (film)",
"Clint Eastwood",
"Sergio Leone"
],
"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 Bridges of Madison County was produced and directed by an American actor who rose to international fame with his role in the Dollars Trilogy, directed by an Italian film director credited as the creator of the Spaghetti Western genre.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Bridges of Madison County (film)",
"Clint Eastwood",
"Sergio Leone"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 630,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
60847
|
United States sanctions against foreign countries until the 19thcentury, includes the countries of China, Lesotho, Afghanistan and Turkmenistan as Tier 3 on Trafficking in Persons Report.
|
{
"evidence": [
"United States sanctions_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:\nUnited States sanctions against foreign countries until the 19thcentury, includes the countries of China, Lesotho, Afghanistan and Turkmenistan as Tier 3 on Trafficking in Persons Report.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"United States sanctions_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 631,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.722",
"value": "?search="
},
{
"operation": "search",
"time": "37.154",
"value": "United States sanctions"
},
{
"operation": "Now on",
"time": "37.915",
"value": "United States sanctions"
},
{
"operation": "Highlighting",
"time": "129.963",
"value": "United States sanctions_sentence_0"
},
{
"operation": "hyperlink",
"time": "184.491",
"value": "Embargo Act of 1807"
},
{
"operation": "Now on",
"time": "185.137",
"value": "Embargo Act of 1807"
},
{
"operation": "finish",
"time": "201.584",
"value": "finish"
}
]
}
|
exfever_8218
|
Taylor Jane Schilling is the most famous American actress.
|
{
"evidence": [
"Taylor Schilling",
"Golden Globe Award for Best Actress – Television Series Musical or Comedy"
],
"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:\nTaylor Jane Schilling is the most famous American actress.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Taylor Schilling",
"Golden Globe Award for Best Actress – Television Series Musical or Comedy"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 632,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
66999
|
The DSM-IV-TR contained four sub-classifications of schizophrenia and the ICD-6, maintained by the World Health Organization (WHO) defines six sub-classifications of schizophrenia.
|
{
"evidence": [
"Diagnosis of schizophrenia_sentence_19"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe DSM-IV-TR contained four sub-classifications of schizophrenia and the ICD-6, maintained by the World Health Organization (WHO) defines six sub-classifications of schizophrenia.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Diagnosis of schizophrenia_sentence_19"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 633,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "3.817",
"value": "?search="
},
{
"operation": "search",
"time": "23.158",
"value": "Schizophrenia"
},
{
"operation": "Now on",
"time": "25.49",
"value": "Schizophrenia"
},
{
"operation": "Page search",
"time": "38.43",
"value": "World Health Organization"
},
{
"operation": "Page search",
"time": "64.006",
"value": "sub-classifications"
},
{
"operation": "Highlighting",
"time": "102.582",
"value": "Schizophrenia_sentence_210"
},
{
"operation": "Highlighting",
"time": "102.584",
"value": "Schizophrenia_title"
},
{
"operation": "Highlighting",
"time": "102.585",
"value": "Schizophrenia_section_10"
},
{
"operation": "Highlighting",
"time": "102.586",
"value": "Schizophrenia_section_12"
},
{
"operation": "Highlighting deleted",
"time": "116.146",
"value": "Schizophrenia_sentence_210"
},
{
"operation": "Highlighting deleted",
"time": "116.149",
"value": "Schizophrenia_title"
},
{
"operation": "Highlighting deleted",
"time": "116.154",
"value": "Schizophrenia_section_10"
},
{
"operation": "Highlighting deleted",
"time": "116.156",
"value": "Schizophrenia_section_12"
},
{
"operation": "hyperlink",
"time": "118.155",
"value": "sub-classifications"
},
{
"operation": "Now on",
"time": "119.352",
"value": "Diagnosis of schizophrenia#Sub-classifications"
},
{
"operation": "Page search",
"time": "130.95",
"value": "sub-classifications"
},
{
"operation": "Page search",
"time": "147.486",
"value": "DSM-IV-TR"
},
{
"operation": "Highlighting",
"time": "169.588",
"value": "Diagnosis of schizophrenia_sentence_19"
},
{
"operation": "finish",
"time": "209.83",
"value": "finish"
}
]
}
|
49492
|
Adults of Eudeilinia luteifera are white, the forewings hve two curved irregular pale buff lines while the hindwings are similar, but the inner line is smaller, and their larvae feed on Cornus florida, also known as the flowering dogwood, a species of flowering tree native to eastern North America and northern Mexico.
|
{
"evidence": [
"Eudeilinia luteifera_sentence_4",
"Eudeilinia luteifera_sentence_5",
"Eudeilinia luteifera_sentence_7",
"Cornus florida_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:\nAdults of Eudeilinia luteifera are white, the forewings hve two curved irregular pale buff lines while the hindwings are similar, but the inner line is smaller, and their larvae feed on Cornus florida, also known as the flowering dogwood, a species of flowering tree native to eastern North America and northern Mexico.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Eudeilinia luteifera_sentence_4",
"Eudeilinia luteifera_sentence_5",
"Eudeilinia luteifera_sentence_7",
"Cornus florida_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 634,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.104",
"value": "?search="
},
{
"operation": "search",
"time": "14.305",
"value": "Eudeilinia luteifera"
},
{
"operation": "Now on",
"time": "16.236",
"value": "Eudeilinia luteifera"
},
{
"operation": "Highlighting",
"time": "22.697",
"value": "Eudeilinia luteifera_sentence_4"
},
{
"operation": "Highlighting",
"time": "32.904",
"value": "Eudeilinia luteifera_sentence_5"
},
{
"operation": "Highlighting",
"time": "43.627",
"value": "Eudeilinia luteifera_sentence_7"
},
{
"operation": "hyperlink",
"time": "49.724",
"value": "Cornus florida"
},
{
"operation": "Now on",
"time": "51.389",
"value": "Cornus florida"
},
{
"operation": "Highlighting",
"time": "53.043",
"value": "Cornus florida_sentence_0"
},
{
"operation": "finish",
"time": "96.019",
"value": "finish"
}
]
}
|
exfever_1290
|
Daniel Michael DeVito Jr. is an American actor known for his film roles in an American family comedy-drama film directed, written, and produced by James L. Brooks.
|
{
"evidence": [
"Danny DeVito",
"Terms of Endearment"
],
"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:\nDaniel Michael DeVito Jr. is an American actor known for his film roles in an American family comedy-drama film directed, written, and produced by James L. Brooks.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Danny DeVito",
"Terms of Endearment"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 635,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
35233
|
Singapore had the most number of athletes participating in cue sports at the Asian Games in 2010 (held in Guangzhou, China).
|
{
"evidence": [
"Cue sports at the 2010 Asian Games_sentence_1",
"Cue sports at the 2010 Asian Games_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:\nSingapore had the most number of athletes participating in cue sports at the Asian Games in 2010 (held in Guangzhou, China).\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Cue sports at the 2010 Asian Games_sentence_1",
"Cue sports at the 2010 Asian Games_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 636,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.654",
"value": "?search="
},
{
"operation": "search",
"time": "58.133",
"value": "Cue sports at the 2010 Asian Games"
},
{
"operation": "Now on",
"time": "59.145",
"value": "Cue sports at the 2010 Asian Games"
},
{
"operation": "Page search",
"time": "61.328",
"value": "Singapore"
},
{
"operation": "Highlighting",
"time": "84.729",
"value": "Cue sports at the 2010 Asian Games_sentence_1"
},
{
"operation": "Highlighting",
"time": "210.28",
"value": "Cue sports at the 2010 Asian Games_cell_3_4_0"
},
{
"operation": "Highlighting",
"time": "210.281",
"value": "Cue sports at the 2010 Asian Games_section_1"
},
{
"operation": "Highlighting",
"time": "210.281",
"value": "Cue sports at the 2010 Asian Games_section_2"
},
{
"operation": "Highlighting",
"time": "210.282",
"value": "Cue sports at the 2010 Asian Games_header_cell_3_0_3"
},
{
"operation": "Highlighting deleted",
"time": "213.961",
"value": "Cue sports at the 2010 Asian Games_cell_3_4_0"
},
{
"operation": "Highlighting deleted",
"time": "213.962",
"value": "Cue sports at the 2010 Asian Games_section_1"
},
{
"operation": "Highlighting deleted",
"time": "213.963",
"value": "Cue sports at the 2010 Asian Games_section_2"
},
{
"operation": "Highlighting deleted",
"time": "213.964",
"value": "Cue sports at the 2010 Asian Games_header_cell_3_0_3"
},
{
"operation": "Highlighting",
"time": "226.561",
"value": "Cue sports at the 2010 Asian Games_sentence_0"
},
{
"operation": "finish",
"time": "230.547",
"value": "finish"
}
]
}
|
93789
|
The Music Association of Ireland was set up in 1948 to improve the position of classical music within the cultural life of Ireland which undertook its greatest challenge to date when it organised a festival of concerts and lectures to commemorate the bicentenary of the death of Johann Sebastian Bach in 1960, where for several months, members raised funds and tried to source Baroque instruments.
|
{
"evidence": [
"Music Association of Ireland_sentence_10"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Music Association of Ireland was set up in 1948 to improve the position of classical music within the cultural life of Ireland which undertook its greatest challenge to date when it organised a festival of concerts and lectures to commemorate the bicentenary of the death of Johann Sebastian Bach in 1960, where for several months, members raised funds and tried to source Baroque instruments.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Music Association of Ireland_sentence_10"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 637,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.905",
"value": "?search="
},
{
"operation": "search",
"time": "19.49",
"value": "Music Association of Ireland"
},
{
"operation": "Now on",
"time": "20.11",
"value": "Music Association of Ireland"
},
{
"operation": "Highlighting",
"time": "34.541",
"value": "Music Association of Ireland_sentence_0"
},
{
"operation": "Highlighting",
"time": "98.781",
"value": "Music Association of Ireland_sentence_10"
},
{
"operation": "Highlighting",
"time": "131.577",
"value": "Music Association of Ireland_sentence_10"
},
{
"operation": "finish",
"time": "166.963",
"value": "finish"
}
]
}
|
exfever_6900
|
The 1972 American science fiction film, directed by J. Lee Thompson, was an adaptation of a science fiction media franchise from 1968.
|
{
"evidence": [
"Conquest of the Planet of the Apes",
"Planet of the Apes"
],
"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 1972 American science fiction film, directed by J. Lee Thompson, was an adaptation of a science fiction media franchise from 1968.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Conquest of the Planet of the Apes",
"Planet of the Apes"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 638,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
45377
|
Laetitia du Couëdic is an equestrian who joined Haras de la Chesnaye in 2020.
|
{
"evidence": [
"Laetitia du Couëdic_sentence_14"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nLaetitia du Couëdic is an equestrian who joined Haras de la Chesnaye in 2020.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Laetitia du Couëdic_sentence_14"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 639,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.501",
"value": "?search="
},
{
"operation": "search",
"time": "18.237",
"value": "Laetitia du Couëdic"
},
{
"operation": "Now on",
"time": "19.74",
"value": "Laetitia du Couëdic"
},
{
"operation": "Highlighting",
"time": "25.843",
"value": "Laetitia du Couëdic_sentence_0"
},
{
"operation": "Highlighting",
"time": "46.861",
"value": "Laetitia du Couëdic_sentence_14"
},
{
"operation": "Highlighting deleted",
"time": "53.796",
"value": "Laetitia du Couëdic_sentence_0"
},
{
"operation": "finish",
"time": "78.541",
"value": "finish"
}
]
}
|
67004
|
The E1 domain contains two independent folding units, GFLD has a highly charged basic surface and is not packed on any organic substance.
|
{
"evidence": [
"APLP2_sentence_5",
"APLP2_sentence_6",
"APLP2_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:\nThe E1 domain contains two independent folding units, GFLD has a highly charged basic surface and is not packed on any organic substance.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"APLP2_sentence_5",
"APLP2_sentence_6",
"APLP2_sentence_7"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 640,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.733",
"value": "?search="
},
{
"operation": "search",
"time": "13.078",
"value": "Contains... E1 domain"
},
{
"operation": "Now on",
"time": "19.275",
"value": "undefined"
},
{
"operation": "search",
"time": "35.471",
"value": "Contains...GFLD"
},
{
"operation": "Now on",
"time": "36.832",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "60.853",
"value": "Amyloid precursor protein"
},
{
"operation": "Now on",
"time": "62.355",
"value": "Amyloid precursor protein"
},
{
"operation": "Page search",
"time": "81.079",
"value": "e1 domain"
},
{
"operation": "hyperlink",
"time": "137.565",
"value": "growth factor-like domain"
},
{
"operation": "Now on",
"time": "138.897",
"value": "Growth factor-like domain"
},
{
"operation": "hyperlink",
"time": "165.502",
"value": "protein domain"
},
{
"operation": "Now on",
"time": "175.527",
"value": "Protein domain"
},
{
"operation": "Page search",
"time": "215.844",
"value": "e1"
},
{
"operation": "search",
"time": "246.912",
"value": "Contains... E1 domain contains two independent folding units, GFLD has a highly charged basic surface"
},
{
"operation": "Now on",
"time": "250.603",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "255.879",
"value": "APLP2"
},
{
"operation": "Now on",
"time": "256.801",
"value": "APLP2"
},
{
"operation": "Highlighting",
"time": "264.437",
"value": "APLP2_sentence_5"
},
{
"operation": "Highlighting",
"time": "269.529",
"value": "APLP2_sentence_6"
},
{
"operation": "Page search",
"time": "293.435",
"value": "organic"
},
{
"operation": "search",
"time": "347.31",
"value": "gfld organic"
},
{
"operation": "Now on",
"time": "348.793",
"value": "undefined"
},
{
"operation": "search",
"time": "364.108",
"value": "gfld organic"
},
{
"operation": "Now on",
"time": "364.68",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "375.555",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "375.564",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "375.662",
"value": "APLP2"
},
{
"operation": "Page search",
"time": "391.1",
"value": "packed"
},
{
"operation": "hyperlink",
"time": "404.86",
"value": "beta-sheet"
},
{
"operation": "Now on",
"time": "405.716",
"value": "Beta sheet"
},
{
"operation": "back-button-clicked",
"time": "423.445",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "423.576",
"value": "APLP2"
},
{
"operation": "Highlighting",
"time": "433.996",
"value": "APLP2_sentence_7"
},
{
"operation": "finish",
"time": "445.881",
"value": "finish"
}
]
}
|
11609
|
John Cougar Concentration Camp, formed in January 1994 after a pun on John Cougar Mellencamp, played their first show at the Casbah (a nightclub near downtown San Diego, California) on January 31, 1994.
|
{
"evidence": [
"Jon Cougar Concentration Camp_sentence_1",
"Jon Cougar Concentration Camp_sentence_2",
"Jon Cougar Concentration Camp_sentence_3",
"The Casbah (music venue)_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:\nJohn Cougar Concentration Camp, formed in January 1994 after a pun on John Cougar Mellencamp, played their first show at the Casbah (a nightclub near downtown San Diego, California) on January 31, 1994.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Jon Cougar Concentration Camp_sentence_1",
"Jon Cougar Concentration Camp_sentence_2",
"Jon Cougar Concentration Camp_sentence_3",
"The Casbah (music venue)_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 641,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.515",
"value": "?search="
},
{
"operation": "search",
"time": "21.278",
"value": "Contains...John Cougar Concentration Camp"
},
{
"operation": "Now on",
"time": "22.289",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "26.969",
"value": "Jon Cougar Concentration Camp"
},
{
"operation": "Now on",
"time": "27.505",
"value": "Jon Cougar Concentration Camp"
},
{
"operation": "Highlighting",
"time": "36.522",
"value": "Jon Cougar Concentration Camp_sentence_1"
},
{
"operation": "Highlighting",
"time": "38.467",
"value": "Jon Cougar Concentration Camp_sentence_2"
},
{
"operation": "Highlighting",
"time": "49.618",
"value": "Jon Cougar Concentration Camp_sentence_3"
},
{
"operation": "hyperlink",
"time": "51.6",
"value": "The Casbah"
},
{
"operation": "Now on",
"time": "52.349",
"value": "The Casbah (music venue)"
},
{
"operation": "Highlighting",
"time": "57.665",
"value": "The Casbah (music venue)_sentence_0"
},
{
"operation": "finish",
"time": "86.65",
"value": "finish"
}
]
}
|
61921
|
YouTube channel How to Cook That is a popular source for learning to bake.
|
{
"evidence": [
"How to Cook That_sentence_0",
"How to Cook That_sentence_18"
],
"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:\nYouTube channel How to Cook That is a popular source for learning to bake.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"How to Cook That_sentence_0",
"How to Cook That_sentence_18"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 642,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.149",
"value": "?search="
},
{
"operation": "search",
"time": "18.559",
"value": "How to Cook That"
},
{
"operation": "Now on",
"time": "21.358",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "22.816",
"value": "How to Cook That"
},
{
"operation": "Now on",
"time": "23.371",
"value": "How to Cook That"
},
{
"operation": "Highlighting",
"time": "25.517",
"value": "How to Cook That_sentence_0"
},
{
"operation": "Page search",
"time": "53.99",
"value": "popul"
},
{
"operation": "Page search",
"time": "54.19",
"value": "popul"
},
{
"operation": "Page search",
"time": "54.341",
"value": "popul"
},
{
"operation": "Highlighting",
"time": "82.028",
"value": "How to Cook That_sentence_7"
},
{
"operation": "Highlighting",
"time": "153.36",
"value": "How to Cook That_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "177.49",
"value": "How to Cook That_sentence_7"
},
{
"operation": "Highlighting",
"time": "264.471",
"value": "How to Cook That_sentence_18"
},
{
"operation": "Highlighting deleted",
"time": "332.215",
"value": "How to Cook That_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "332.664",
"value": "How to Cook That_sentence_0"
},
{
"operation": "Highlighting",
"time": "389.434",
"value": "How to Cook That_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "392.123",
"value": "How to Cook That_sentence_18"
},
{
"operation": "Highlighting",
"time": "392.647",
"value": "How to Cook That_sentence_18"
},
{
"operation": "Page search",
"time": "412.55",
"value": "learn"
},
{
"operation": "Page search",
"time": "412.758",
"value": "learn"
},
{
"operation": "Highlighting deleted",
"time": "423.119",
"value": "How to Cook That_sentence_18"
},
{
"operation": "Highlighting",
"time": "424.439",
"value": "How to Cook That_sentence_18"
},
{
"operation": "finish",
"time": "435.73",
"value": "finish"
}
]
}
|
exfever_586
|
Inside Man is a 2006 American heist thriller film whose stars include an American actress.
|
{
"evidence": [
"Inside Man",
"Jodie Foster"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nInside Man is a 2006 American heist thriller film whose stars include an American actress.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Inside Man",
"Jodie Foster"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 643,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
78418
|
The Tokyo Metropolitan University is a Public Research University in Japan which receives public funds through a national or subnational government, as opposed to a private university.
|
{
"evidence": [
"Tokyo Metropolitan University_sentence_0",
"Public university_sentence_1"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Tokyo Metropolitan University is a Public Research University in Japan which receives public funds through a national or subnational government, as opposed to a private university.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Tokyo Metropolitan University_sentence_0",
"Public university_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 644,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.885",
"value": "?search="
},
{
"operation": "search",
"time": "24.207",
"value": "Tokyo Metropolitan University"
},
{
"operation": "Now on",
"time": "25.467",
"value": "Tokyo Metropolitan University"
},
{
"operation": "Highlighting",
"time": "90.481",
"value": "Tokyo Metropolitan University_sentence_0"
},
{
"operation": "hyperlink",
"time": "94.218",
"value": "Public Research University"
},
{
"operation": "Now on",
"time": "96.304",
"value": "Public university"
},
{
"operation": "Highlighting",
"time": "98.544",
"value": "Public university_sentence_1"
},
{
"operation": "finish",
"time": "138.127",
"value": "finish"
}
]
}
|
exfever_4632
|
Another One Bites the Dust is a movie by the British rock band Queen that reached number seven on listing the top-selling singles in the United Kingdom-based on physical sales, paid-for downloads, and streaming.
|
{
"evidence": [
"Another One Bites the Dust",
"UK Singles Chart"
],
"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:\nAnother One Bites the Dust is a movie by the British rock band Queen that reached number seven on listing the top-selling singles in the United Kingdom-based on physical sales, paid-for downloads, and streaming.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Another One Bites the Dust",
"UK Singles Chart"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 645,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3667
|
Richard Dreyfuss starred in a 1980 American coming-of-age story directed by Rob Reiner.
|
{
"evidence": [
"Richard Dreyfuss",
"Stand by Me (film)"
],
"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:\nRichard Dreyfuss starred in a 1980 American coming-of-age story directed by Rob Reiner.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Richard Dreyfuss",
"Stand by Me (film)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 646,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3766
|
50 First Dates is a 2003 American romantic comedy film and was shot on location in the third-largest of the Hawaiian Islands and is within the capital and largest city of the U.S. state of Hawaii.
|
{
"evidence": [
"50 First Dates",
"Oahu",
"Honolulu"
],
"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 First Dates is a 2003 American romantic comedy film and was shot on location in the third-largest of the Hawaiian Islands and is within the capital and largest city of the U.S. state of Hawaii.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"50 First Dates",
"Oahu",
"Honolulu"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 647,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_784
|
David Dhawan is best known for directing several successful films, including a 1980 Indian Hindi-language drama film starring an Indian actor, comedian, dancer, and former politician known for his work in Hindi-language films.
|
{
"evidence": [
"David Dhawan",
"Swarg",
"Govinda (actor)"
],
"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:\nDavid Dhawan is best known for directing several successful films, including a 1980 Indian Hindi-language drama film starring an Indian actor, comedian, dancer, and former politician known for his work in Hindi-language films.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"David Dhawan",
"Swarg",
"Govinda (actor)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 648,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1639
|
The Bible is a collection of religuous texts, originally written in a semitic languages spoken by the israelites.
|
{
"evidence": [
"Bible",
"Biblical Hebrew"
],
"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 Bible is a collection of religuous texts, originally written in a semitic languages spoken by the israelites.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Bible",
"Biblical Hebrew"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 649,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5465
|
Grunge emerged in Washington, whose meaning refers to musicians influenced by the musical style of the late 1970s that?typically produce short, fast-paced songs with hard-edged melodies.
|
{
"evidence": [
"Grunge",
"Alternative rock",
"Punk rock"
],
"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:\nGrunge emerged in Washington, whose meaning refers to musicians influenced by the musical style of the late 1970s that?typically produce short, fast-paced songs with hard-edged melodies.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Grunge",
"Alternative rock",
"Punk rock"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 650,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8212
|
Coldplay drummer is an American musician and songwriter.
|
{
"evidence": [
"Coldplay",
"Will Champion"
],
"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:\nColdplay drummer is an American musician and songwriter.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Coldplay",
"Will Champion"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 651,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5840
|
Dreamer is a 2005 American sports drama film written and directed by John Gatins and starred who received an award nomination at the age of eight given by the Screen Actors Guild-American Federation of Television and Radio Artists.
|
{
"evidence": [
"Dreamer (2005 film)",
"Dakota Fanning",
"Screen Actors Guild 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:\nDreamer is a 2005 American sports drama film written and directed by John Gatins and starred who received an award nomination at the age of eight given by the Screen Actors Guild-American Federation of Television and Radio Artists.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Dreamer (2005 film)",
"Dakota Fanning",
"Screen Actors Guild Award"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 652,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5134
|
The New York Times is an American daily newspaper based in New Jersey City, founded by an American journalist and politician.
|
{
"evidence": [
"The New York Times",
"Henry Jarvis Raymond"
],
"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 New York Times is an American daily newspaper based in New Jersey City, founded by an American journalist and politician.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The New York Times",
"Henry Jarvis Raymond"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 653,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
55697
|
India is Vanuatu's 5th largest export destination, with US$5.5 million worth of goods sent in the 2013-2014 year; which makes up 1.0% of all of Vanuatu's exports and is an increase from US$5.37 million in the 2012-2013 year.
|
{
"evidence": [
"India–Vanuatu relations_sentence_14",
"India–Vanuatu relations_sentence_15"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nIndia is Vanuatu's 5th largest export destination, with US$5.5 million worth of goods sent in the 2013-2014 year; which makes up 1.0% of all of Vanuatu's exports and is an increase from US$5.37 million in the 2012-2013 year.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"India–Vanuatu relations_sentence_14",
"India–Vanuatu relations_sentence_15"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 654,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.84",
"value": "?search="
},
{
"operation": "search",
"time": "21.624",
"value": "Vanuatu"
},
{
"operation": "Now on",
"time": "23.559",
"value": "Vanuatu"
},
{
"operation": "hyperlink",
"time": "31.608",
"value": "5 Economy Vanuatu_section_18"
},
{
"operation": "Page search",
"time": "70.46",
"value": "EXPORT*"
},
{
"operation": "hyperlink",
"time": "147.355",
"value": "Index of Vanuatu-related articles"
},
{
"operation": "Now on",
"time": "148.616",
"value": "Index of Vanuatu-related articles"
},
{
"operation": "hyperlink",
"time": "156.484",
"value": "5 E Index of Vanuatu-related articles_section_4"
},
{
"operation": "hyperlink",
"time": "184.665",
"value": "Vanuatu Council of Trade Unions"
},
{
"operation": "Now on",
"time": "185.699",
"value": "Vanuatu Council of Trade Unions"
},
{
"operation": "search",
"time": "197.535",
"value": "VANUATU EXPORTS"
},
{
"operation": "Now on",
"time": "199.169",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "204.153",
"value": "Economy of Vanuatu"
},
{
"operation": "Now on",
"time": "205.078",
"value": "Economy of Vanuatu"
},
{
"operation": "hyperlink",
"time": "209.93",
"value": "1.6 Exports Economy of Vanuatu_section_6"
},
{
"operation": "hyperlink",
"time": "230.08",
"value": "2 Economic statistics Economy of Vanuatu_section_9"
},
{
"operation": "search",
"time": "274.134",
"value": "VANUATU INDIA"
},
{
"operation": "Now on",
"time": "275.561",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "278.975",
"value": "India–Vanuatu relations"
},
{
"operation": "Now on",
"time": "279.339",
"value": "India–Vanuatu relations"
},
{
"operation": "Highlighting",
"time": "291.582",
"value": "India–Vanuatu relations_sentence_14"
},
{
"operation": "Highlighting",
"time": "296.175",
"value": "India–Vanuatu relations_sentence_15"
},
{
"operation": "finish",
"time": "329.709",
"value": "finish"
}
]
}
|
exfever_1434
|
Wonder Woman is a superheroine fictional character who sometimes adopts her civilian identity as a WAVES yeoman in the 1920s.
|
{
"evidence": [
"Wonder Woman",
"Diana Prince"
],
"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:\nWonder Woman is a superheroine fictional character who sometimes adopts her civilian identity as a WAVES yeoman in the 1920s.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Wonder Woman",
"Diana Prince"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 655,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6159
|
The Girl on the Train is a 2016 mystery psychological thriller film whose genre combines the thriller and psychological fiction genres.
|
{
"evidence": [
"The Girl on the Train (2016 film)",
"Psychological thriller"
],
"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 Girl on the Train is a 2016 mystery psychological thriller film whose genre combines the thriller and psychological fiction genres.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Girl on the Train (2016 film)",
"Psychological thriller"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 656,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7400
|
France is the richest country to attend the 19th G7 Summit.
|
{
"evidence": [
"19th G7 summit",
"Italy"
],
"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:\nFrance is the richest country to attend the 19th G7 Summit.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"19th G7 summit",
"Italy"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 657,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7488
|
Caitlyn Marie Jenner is a retired Olympic gold medal-winning combined event in athletics that consists of ten track and field events, that were also played in the 1914 Summer Olympics, an international multi-sport event held in Stockholm, Sweden.
|
{
"evidence": [
"Caitlyn Jenner",
"Decathlon",
"1912 Summer Olympics"
],
"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:\nCaitlyn Marie Jenner is a retired Olympic gold medal-winning combined event in athletics that consists of ten track and field events, that were also played in the 1914 Summer Olympics, an international multi-sport event held in Stockholm, Sweden.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Caitlyn Jenner",
"Decathlon",
"1912 Summer Olympics"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 658,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
78550
|
Gulustan Mahmood Leso took a silver medal in the women's 400 meter race at the 2009 Asian Indoor Games in Hanoi, Vietnam.
|
{
"evidence": [
"Gulustan Mahmood_sentence_5",
"Indoor athletics at the 2009 Asian Indoor Games_sentence_0"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nGulustan Mahmood Leso took a silver medal in the women's 400 meter race at the 2009 Asian Indoor Games in Hanoi, Vietnam.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Gulustan Mahmood_sentence_5",
"Indoor athletics at the 2009 Asian Indoor Games_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 659,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.918",
"value": "?search="
},
{
"operation": "search",
"time": "6.743",
"value": "Gulustan Mahmood Leso"
},
{
"operation": "Now on",
"time": "7.961",
"value": "undefined"
},
{
"operation": "search",
"time": "53.501",
"value": "Gulustan Leso"
},
{
"operation": "Now on",
"time": "54.448",
"value": "undefined"
},
{
"operation": "search",
"time": "60.22",
"value": "Gulustan Mahmood"
},
{
"operation": "Now on",
"time": "61.145",
"value": "Gulustan Mahmood"
},
{
"operation": "Highlighting",
"time": "86.279",
"value": "Gulustan Mahmood_sentence_5"
},
{
"operation": "hyperlink",
"time": "91.684",
"value": "2009 Asian Indoor Games"
},
{
"operation": "Now on",
"time": "94.655",
"value": "Indoor athletics at the 2009 Asian Indoor Games"
},
{
"operation": "Highlighting",
"time": "94.717",
"value": "Indoor athletics at the 2009 Asian Indoor Games_sentence_0"
},
{
"operation": "finish",
"time": "100.118",
"value": "finish"
}
]
}
|
72272
|
Just over ten miles long, Maryland Route 509 is found entirely in Port Republic, Calvert County, and was constructed from east of MD 765 in the early 1930's, and it was extended west to MD two and MD four in the early 80's
|
{
"evidence": [
"Maryland Route 509_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:\nJust over ten miles long, Maryland Route 509 is found entirely in Port Republic, Calvert County, and was constructed from east of MD 765 in the early 1930's, and it was extended west to MD two and MD four in the early 80's\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Maryland Route 509_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 660,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.393",
"value": "?search="
},
{
"operation": "search",
"time": "12.332",
"value": "Maryland Route 509"
},
{
"operation": "Now on",
"time": "13.109",
"value": "Maryland Route 509"
},
{
"operation": "Highlighting",
"time": "24.594",
"value": "Maryland Route 509_sentence_2"
},
{
"operation": "finish",
"time": "150.475",
"value": "finish"
}
]
}
|
exfever_1243
|
Shahid Kapoor is an Indian actor who appears in Hindi films and trained as a dancer at the academy of an Indian choreographer who is known as the guru of contemporary dance in Pakistan
|
{
"evidence": [
"Shahid Kapoor",
"Shiamak Davar"
],
"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:\nShahid Kapoor is an Indian actor who appears in Hindi films and trained as a dancer at the academy of an Indian choreographer who is known as the guru of contemporary dance in Pakistan\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Shahid Kapoor",
"Shiamak Davar"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 661,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3170
|
Saif Ali Khan is an Indian actor who starred in the 1994 Indian action-comedy film directed by Sameer Malkan.
|
{
"evidence": [
"Saif Ali Khan",
"Main Khiladi Tu Anari"
],
"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:\nSaif Ali Khan is an Indian actor who starred in the 1994 Indian action-comedy film directed by Sameer Malkan.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Saif Ali Khan",
"Main Khiladi Tu Anari"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 662,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8875
|
Theodore Roosevelt Jr. assumed the presidency after the 20th president of the United States, was shot on the grounds of the Pan-American Exposition.
|
{
"evidence": [
"Theodore Roosevelt",
"Assassination of William McKinley"
],
"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:\nTheodore Roosevelt Jr. assumed the presidency after the 20th president of the United States, was shot on the grounds of the Pan-American Exposition.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Theodore Roosevelt",
"Assassination of William McKinley"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 663,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
61088
|
During Good Friday of the Holy Week (which is, the week immediately before Easter), some parishes of the Anglican, Catholic, and Lutheran Church the "Two Hours Devotion" is observed in which this traditionally consists of a series of sermons, interspersed with singing, one on each of the Seven Last Words from the Cross, together with an introduction and a conclusion.
|
{
"evidence": [
"Three Hours' Agony_sentence_0",
"Three Hours' Agony_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:\nDuring Good Friday of the Holy Week (which is, the week immediately before Easter), some parishes of the Anglican, Catholic, and Lutheran Church the \"Two Hours Devotion\" is observed in which this traditionally consists of a series of sermons, interspersed with singing, one on each of the Seven Last Words from the Cross, together with an introduction and a conclusion.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Three Hours' Agony_sentence_0",
"Three Hours' Agony_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 664,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.63",
"value": "?search="
},
{
"operation": "search",
"time": "26.999",
"value": "two hours of devotion"
},
{
"operation": "Now on",
"time": "28.928",
"value": "undefined"
},
{
"operation": "search",
"time": "63.987",
"value": "Good Friday"
},
{
"operation": "Now on",
"time": "66.183",
"value": "Good Friday"
},
{
"operation": "Highlighting",
"time": "83.109",
"value": "Good Friday_sentence_4"
},
{
"operation": "Highlighting",
"time": "83.111",
"value": "Good Friday_title"
},
{
"operation": "hyperlink",
"time": "91.55",
"value": "Holy Week"
},
{
"operation": "Now on",
"time": "93.359",
"value": "Holy Week"
},
{
"operation": "Highlighting",
"time": "98.32",
"value": "Holy Week_sentence_5"
},
{
"operation": "Highlighting",
"time": "98.321",
"value": "Holy Week_title"
},
{
"operation": "back-button-clicked",
"time": "99.785",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "99.908",
"value": "Good Friday"
},
{
"operation": "hyperlink",
"time": "125.826",
"value": "5 Anglican Communion Good Friday_section_17"
},
{
"operation": "Highlighting",
"time": "141.224",
"value": "Good Friday_sentence_169"
},
{
"operation": "Highlighting",
"time": "141.226",
"value": "Good Friday_section_18"
},
{
"operation": "Highlighting deleted",
"time": "191.076",
"value": "Good Friday_sentence_169"
},
{
"operation": "Highlighting deleted",
"time": "191.079",
"value": "Good Friday_section_18"
},
{
"operation": "hyperlink",
"time": "259.339",
"value": "4.3 Three Hours' Agony Good Friday_section_12"
},
{
"operation": "Highlighting",
"time": "267.759",
"value": "Good Friday_sentence_135"
},
{
"operation": "Highlighting",
"time": "267.761",
"value": "Good Friday_section_9"
},
{
"operation": "Highlighting",
"time": "267.761",
"value": "Good Friday_section_12"
},
{
"operation": "hyperlink",
"time": "268.542",
"value": "Three Hours' Agony"
},
{
"operation": "Now on",
"time": "269.901",
"value": "Three Hours' Agony"
},
{
"operation": "Highlighting",
"time": "273.762",
"value": "Three Hours' Agony_sentence_0"
},
{
"operation": "Highlighting",
"time": "293.715",
"value": "Three Hours' Agony_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "312.592",
"value": "Three Hours' Agony_sentence_1"
},
{
"operation": "Highlighting",
"time": "313.273",
"value": "Three Hours' Agony_sentence_1"
},
{
"operation": "hyperlink",
"time": "321.03",
"value": "sayings of Jesus on the cross"
},
{
"operation": "Now on",
"time": "322.378",
"value": "Sayings of Jesus on the cross"
},
{
"operation": "finish",
"time": "425.382",
"value": "finish"
}
]
}
|
exfever_4266
|
Natalie Wood was an American actress and model who starred in John Ford's 1956 American Technicolor VistaVision epic film, whose genre of fiction and film is set primarily in the latter half of the early 18th century in the Western United States.
|
{
"evidence": [
"Natalie Wood",
"The Searchers",
"Western (genre)"
],
"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:\nNatalie Wood was an American actress and model who starred in John Ford's 1956 American Technicolor VistaVision epic film, whose genre of fiction and film is set primarily in the latter half of the early 18th century in the Western United States.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Natalie Wood",
"The Searchers",
"Western (genre)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 665,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
38636
|
A Sinclair user admitted the ZX-81, which was developed by Sinclair Research before launching in 1981, should be recognized as the first graphics adventure while another Sinclair user provided positive reviews for the game.
|
{
"evidence": [
"ZX81_sentence_213",
"ZX81_sentence_0",
"ZX81_sentence_1",
"ZX81_sentence_269",
"ZX81_sentence_270",
"ZX81_sentence_273"
],
"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:\nA Sinclair user admitted the ZX-81, which was developed by Sinclair Research before launching in 1981, should be recognized as the first graphics adventure while another Sinclair user provided positive reviews for the game.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"ZX81_sentence_213",
"ZX81_sentence_0",
"ZX81_sentence_1",
"ZX81_sentence_269",
"ZX81_sentence_270",
"ZX81_sentence_273"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 666,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.127",
"value": "?search="
},
{
"operation": "search",
"time": "15.088",
"value": "Zx-81"
},
{
"operation": "Now on",
"time": "16.563",
"value": "ZX81"
},
{
"operation": "Highlighting",
"time": "118.017",
"value": "ZX81_sentence_213"
},
{
"operation": "Highlighting",
"time": "186.383",
"value": "ZX81_sentence_0"
},
{
"operation": "Highlighting",
"time": "189.06",
"value": "ZX81_sentence_1"
},
{
"operation": "Highlighting",
"time": "309.304",
"value": "ZX81_sentence_15ZX81_sentence_16"
},
{
"operation": "Highlighting",
"time": "333.754",
"value": "ZX81_sentence_269"
},
{
"operation": "Highlighting",
"time": "336.386",
"value": "ZX81_sentence_270"
},
{
"operation": "Highlighting",
"time": "348.491",
"value": "ZX81_sentence_273"
},
{
"operation": "finish",
"time": "403.898",
"value": "finish"
}
]
}
|
48294
|
Pirelli Internetional Award is the first international multimedia competition sponsored by the Pirelli Corporation, the 7th-largest tyre manufacturer.
|
{
"evidence": [
"Pirelli Internetional Award_sentence_0",
"Pirelli Internetional Award_sentence_4",
"Pirelli_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:\nPirelli Internetional Award is the first international multimedia competition sponsored by the Pirelli Corporation, the 7th-largest tyre manufacturer.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Pirelli Internetional Award_sentence_0",
"Pirelli Internetional Award_sentence_4",
"Pirelli_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 667,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.734",
"value": "?search="
},
{
"operation": "search",
"time": "10.998",
"value": "Pirelli Internetional Award"
},
{
"operation": "Now on",
"time": "11.707",
"value": "Pirelli Internetional Award"
},
{
"operation": "Highlighting",
"time": "20.607",
"value": "Pirelli Internetional Award_sentence_0"
},
{
"operation": "Highlighting",
"time": "26.297",
"value": "Pirelli Internetional Award_sentence_4"
},
{
"operation": "hyperlink",
"time": "31.96",
"value": "Pirelli Corporation"
},
{
"operation": "Now on",
"time": "32.728",
"value": "Pirelli"
},
{
"operation": "Highlighting",
"time": "36.641",
"value": "Pirelli_sentence_1"
},
{
"operation": "finish",
"time": "44.813",
"value": "finish"
}
]
}
|
7863
|
Jožka Řehová, born on April 16, 1896 in Vítkovice (an administrative district of the city of Ostrava), worked in Vítkovice Ironworks and later in the company's hospital during the First World War.
|
{
"evidence": [
"Jožka Jabůrková_sentence_3",
"Jožka Jabůrková_sentence_4",
"Vítkovice (Ostrava)_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:\nJožka Řehová, born on April 16, 1896 in Vítkovice (an administrative district of the city of Ostrava), worked in Vítkovice Ironworks and later in the company's hospital during the First World War.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Jožka Jabůrková_sentence_3",
"Jožka Jabůrková_sentence_4",
"Vítkovice (Ostrava)_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 668,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.555",
"value": "?search="
},
{
"operation": "search",
"time": "9.132",
"value": "Contains...Jožka Řehová"
},
{
"operation": "Now on",
"time": "9.857",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "20.117",
"value": "Jožka Jabůrková"
},
{
"operation": "Now on",
"time": "20.654",
"value": "Jožka Jabůrková"
},
{
"operation": "Highlighting",
"time": "28.128",
"value": "Jožka Jabůrková_sentence_3"
},
{
"operation": "Highlighting",
"time": "36.399",
"value": "Jožka Jabůrková_sentence_4"
},
{
"operation": "hyperlink",
"time": "40.806",
"value": "Vítkovice"
},
{
"operation": "Now on",
"time": "41.338",
"value": "Vítkovice (Ostrava)"
},
{
"operation": "Highlighting",
"time": "43.693",
"value": "Vítkovice (Ostrava)_sentence_0"
},
{
"operation": "finish",
"time": "63.342",
"value": "finish"
}
]
}
|
exfever_4328
|
Plato was a Greek philosopher born in Athens during the Classical period in a northeastern Mediterranean civilization, which existed from the 12th–9th centuries BC to the end of classical antiquity.
|
{
"evidence": [
"Plato",
"Ancient Greece"
],
"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:\nPlato was a Greek philosopher born in Athens during the Classical period in a northeastern Mediterranean civilization, which existed from the 12th–9th centuries BC to the end of classical antiquity.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Plato",
"Ancient Greece"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 669,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
47386
|
Michael Elmhirst Cates, the 19th Lucasian Professor of Mathematics at the University of Cambridge, proposed that jammed systems differ from ordinary solids in that if the direction of the applied stress changes, the jam will break up; and sometimes, the change of direction required is very small.
|
{
"evidence": [
"Michael Cates_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:\nMichael Elmhirst Cates, the 19th Lucasian Professor of Mathematics at the University of Cambridge, proposed that jammed systems differ from ordinary solids in that if the direction of the applied stress changes, the jam will break up; and sometimes, the change of direction required is very small.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Michael Cates_sentence_1"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 670,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.13",
"value": "?search="
},
{
"operation": "search",
"time": "17.316",
"value": "Michael Elmhirst Cates"
},
{
"operation": "Now on",
"time": "18.277",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "21.45",
"value": "Michael Cates"
},
{
"operation": "Now on",
"time": "22.141",
"value": "Michael Cates"
},
{
"operation": "Highlighting",
"time": "28.022",
"value": "Michael Cates_sentence_1"
},
{
"operation": "hyperlink",
"time": "66.194",
"value": "constitutive equation"
},
{
"operation": "Now on",
"time": "67.154",
"value": "Constitutive equation"
},
{
"operation": "back-button-clicked",
"time": "84.882",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "84.95",
"value": "Michael Cates"
},
{
"operation": "hyperlink",
"time": "120.612",
"value": "soft matter"
},
{
"operation": "Now on",
"time": "121.138",
"value": "Soft matter"
},
{
"operation": "search",
"time": "135.707",
"value": "jammed systems"
},
{
"operation": "Now on",
"time": "137.062",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "143.561",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "143.63",
"value": "Soft matter"
},
{
"operation": "hyperlink",
"time": "172.731",
"value": "Active matter"
},
{
"operation": "Now on",
"time": "173.237",
"value": "Active matter"
},
{
"operation": "back-button-clicked",
"time": "184.628",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "184.694",
"value": "Soft matter"
},
{
"operation": "back-button-clicked",
"time": "185.557",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "185.571",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "185.577",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "185.643",
"value": "Michael Cates"
},
{
"operation": "hyperlink",
"time": "194.386",
"value": "FRS"
},
{
"operation": "Now on",
"time": "195.151",
"value": "Fellow of the Royal Society"
},
{
"operation": "back-button-clicked",
"time": "219.783",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "219.854",
"value": "Michael Cates"
},
{
"operation": "hyperlink",
"time": "223.816",
"value": "FRSE"
},
{
"operation": "Now on",
"time": "224.341",
"value": "Fellowship of the Royal Society of Edinburgh"
},
{
"operation": "back-button-clicked",
"time": "237.392",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "237.488",
"value": "Michael Cates"
},
{
"operation": "back-button-clicked",
"time": "245.619",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "245.633",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "245.638",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "245.645",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "245.655",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "246.384",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "251.859",
"value": "List of fellows of the Royal Society A, B, C"
},
{
"operation": "Now on",
"time": "256.812",
"value": "List of fellows of the Royal Society A, B, C"
},
{
"operation": "back-button-clicked",
"time": "274.988",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "275.809",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "287.089",
"value": "Michael Cates"
},
{
"operation": "Now on",
"time": "287.445",
"value": "Michael Cates"
},
{
"operation": "hyperlink",
"time": "315.431",
"value": "EPSRC"
},
{
"operation": "Now on",
"time": "316.226",
"value": "Engineering and Physical Sciences Research Council"
},
{
"operation": "finish",
"time": "358.157",
"value": "finish"
}
]
}
|
exfever_6344
|
To Kill a Mockingbird is a 1962 American film directed by Robert Mulligan, whose category of narrative fiction is intended to be more serious than humorous in tone.
|
{
"evidence": [
"To Kill a Mockingbird (film)",
"Drama (film and television)"
],
"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:\nTo Kill a Mockingbird is a 1962 American film directed by Robert Mulligan, whose category of narrative fiction is intended to be more serious than humorous in tone.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"To Kill a Mockingbird (film)",
"Drama (film and television)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 671,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
24221
|
The Yoshizawa–Randlett system (a system used to describe the folds of origami models) was first described in Samuel Randlett's Art of Origami in 1960 after Samuel Randlett and Robert Harbin added a few symbols to Akira Yoshizawa's diagramming.
|
{
"evidence": [
"Yoshizawa–Randlett system_sentence_13"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Yoshizawa–Randlett system (a system used to describe the folds of origami models) was first described in Samuel Randlett's Art of Origami in 1960 after Samuel Randlett and Robert Harbin added a few symbols to Akira Yoshizawa's diagramming.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Yoshizawa–Randlett system_sentence_13"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 672,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.058",
"value": "?search="
},
{
"operation": "search",
"time": "18.258",
"value": "Yoshizawa–Randlett system"
},
{
"operation": "Now on",
"time": "19.165",
"value": "Yoshizawa–Randlett system"
},
{
"operation": "Highlighting",
"time": "27.582",
"value": "Yoshizawa–Randlett system_sentence_0"
},
{
"operation": "Highlighting",
"time": "61.23",
"value": "Yoshizawa–Randlett system_sentence_13"
},
{
"operation": "Highlighting deleted",
"time": "63.337",
"value": "Yoshizawa–Randlett system_sentence_0"
},
{
"operation": "finish",
"time": "69.79",
"value": "finish"
}
]
}
|
64458
|
The curling club, Soo Curler's Association in Ontario, Canada is home to Team Brad Jacobs, an Olympic champion who won gold in the 2014 Winter Olympics, and the 2013 Tim Hortons Brier championship, and is currently a marketing director for World Financial Group.
|
{
"evidence": [
"Brad Jacobs (curler)_sentence_4",
"Brad Jacobs (curler)_sentence_3",
"Brad Jacobs (curler)_sentence_64",
"Soo Curlers Association_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:\nThe curling club, Soo Curler's Association in Ontario, Canada is home to Team Brad Jacobs, an Olympic champion who won gold in the 2014 Winter Olympics, and the 2013 Tim Hortons Brier championship, and is currently a marketing director for World Financial Group.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Brad Jacobs (curler)_sentence_4",
"Brad Jacobs (curler)_sentence_3",
"Brad Jacobs (curler)_sentence_64",
"Soo Curlers Association_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 673,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.979",
"value": "?search="
},
{
"operation": "search",
"time": "80.818",
"value": "Soo Curler's Association"
},
{
"operation": "Now on",
"time": "82.386",
"value": "undefined"
},
{
"operation": "search",
"time": "118.51",
"value": "Brad Jacobs (curler)"
},
{
"operation": "Now on",
"time": "145.243",
"value": "Brad Jacobs (curler)"
},
{
"operation": "Highlighting",
"time": "170.672",
"value": "Brad Jacobs (curler)_sentence_3"
},
{
"operation": "Highlighting",
"time": "180.452",
"value": "Brad Jacobs (curler)_sentence_4"
},
{
"operation": "Highlighting",
"time": "198.177",
"value": "Brad Jacobs (curler)_sentence_3"
},
{
"operation": "Highlighting",
"time": "202.409",
"value": "Brad Jacobs (curler)_sentence_64"
},
{
"operation": "search",
"time": "284.687",
"value": "soo curlers association"
},
{
"operation": "Now on",
"time": "286.186",
"value": "Soo Curlers Association"
},
{
"operation": "Highlighting",
"time": "288.624",
"value": "Soo Curlers Association_sentence_3"
},
{
"operation": "finish",
"time": "294.105",
"value": "finish"
}
]
}
|
exfever_6126
|
Romney was the first Republican nominee in the 1994 United States Senate election in Massachusetts.
|
{
"evidence": [
"Mitt Romney",
"Republican Party (United States)"
],
"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:\nRomney was the first Republican nominee in the 1994 United States Senate election in Massachusetts.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Mitt Romney",
"Republican Party (United States)"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 674,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8827
|
Saint Peter was one of the Twelve Apostles of Jesus Christ who was crucified in Rome under the last emperor of the Julio-Claudian dynasty, reigning from AD 54 until his suicide in 68.
|
{
"evidence": [
"Saint Peter",
"Nero"
],
"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:\nSaint Peter was one of the Twelve Apostles of Jesus Christ who was crucified in Rome under the last emperor of the Julio-Claudian dynasty, reigning from AD 54 until his suicide in 68.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Saint Peter",
"Nero"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 675,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_722
|
England is the largest country in the United Kingdom.
|
{
"evidence": [
"Leyton Orient F.C.",
"London"
],
"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:\nEngland is the largest country in the United Kingdom.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Leyton Orient F.C.",
"London"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 676,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
93109
|
Gangavalli is a town in Salem district in the state of Tamil Nadu, India that is flat.
|
{
"evidence": [
"Gangavalli_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:\nGangavalli is a town in Salem district in the state of Tamil Nadu, India that is flat.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Gangavalli_sentence_1"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 677,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "9.468",
"value": "?search="
},
{
"operation": "search",
"time": "92.815",
"value": "Gangavalli"
},
{
"operation": "Now on",
"time": "95.629",
"value": "Gangavalli"
},
{
"operation": "Highlighting",
"time": "97.025",
"value": "Gangavalli_sentence_1"
},
{
"operation": "Highlighting",
"time": "124.874",
"value": "Gangavalli_sentence_1"
},
{
"operation": "finish",
"time": "211.168",
"value": "finish"
}
]
}
|
exfever_314
|
The Taj Mahal was commissioned by the Mughal emperor Shah Jahan to house the tomb of his favourite wife, which is the centrepiece of the complex that includes a place of worship for Christians.
|
{
"evidence": [
"Taj Mahal",
"Mosque"
],
"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 Taj Mahal was commissioned by the Mughal emperor Shah Jahan to house the tomb of his favourite wife, which is the centrepiece of the complex that includes a place of worship for Christians.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Taj Mahal",
"Mosque"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 678,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6492
|
Brazzers is the only pornographic video production company in the world.
|
{
"evidence": [
"Brazzers",
"High-definition video"
],
"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:\nBrazzers is the only pornographic video production company in the world.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Brazzers",
"High-definition video"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 679,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
19199
|
Ernst Höpfner was appointed Provinzialschulrat in Koblenz, then he served as a curator at the University of Göttingen (a private research university founded in 1734).
|
{
"evidence": [
"University of Göttingen_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:\nErnst Höpfner was appointed Provinzialschulrat in Koblenz, then he served as a curator at the University of Göttingen (a private research university founded in 1734).\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"University of Göttingen_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 680,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.802",
"value": "?search="
},
{
"operation": "search",
"time": "20.408",
"value": "Ernst Höpfner"
},
{
"operation": "Now on",
"time": "21.326",
"value": "Ernst Höpfner"
},
{
"operation": "Highlighting",
"time": "33.558",
"value": "Ernst Höpfner_sentence_3"
},
{
"operation": "Highlighting",
"time": "33.559",
"value": "Ernst Höpfner_title"
},
{
"operation": "Highlighting",
"time": "33.56",
"value": "Ernst Höpfner_section_0"
},
{
"operation": "Highlighting",
"time": "43.009",
"value": "Ernst Höpfner_sentence_4"
},
{
"operation": "hyperlink",
"time": "44.169",
"value": "University of Göttingen"
},
{
"operation": "Now on",
"time": "45.305",
"value": "University of Göttingen"
},
{
"operation": "Highlighting",
"time": "56.605",
"value": "University of Göttingen_sentence_0"
},
{
"operation": "finish",
"time": "65.705",
"value": "finish"
}
]
}
|
70508
|
Kam Wong not only argues that the dynamics of imperial China form the basis for contemporary Chinese understandings of terrorism, but also argues that fear of chaos and social disorder is a powerful factor in mobilizing political will to combat potential threats.
|
{
"evidence": [
"Terrorism in China_sentence_18",
"Terrorism in China_sentence_19"
],
"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:\nKam Wong not only argues that the dynamics of imperial China form the basis for contemporary Chinese understandings of terrorism, but also argues that fear of chaos and social disorder is a powerful factor in mobilizing political will to combat potential threats.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Terrorism in China_sentence_18",
"Terrorism in China_sentence_19"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 681,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.681",
"value": "?search="
},
{
"operation": "search",
"time": "19.278",
"value": "Kam Wong"
},
{
"operation": "Now on",
"time": "21.555",
"value": "undefined"
},
{
"operation": "search",
"time": "45.619",
"value": "ear of chaos and social disorder is a powerful factor"
},
{
"operation": "Now on",
"time": "46.915",
"value": "undefined"
},
{
"operation": "search",
"time": "59.995",
"value": "powerful factor in mobilizing political will to combat potential threats"
},
{
"operation": "Now on",
"time": "62.308",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "65.712",
"value": "Regionally focused terrorism and political violence Terrorism in China_section_1"
},
{
"operation": "Now on",
"time": "66.543",
"value": "Terrorism in China#Regionally focused terrorism and political violence Terrorism in China section 1"
},
{
"operation": "Page search",
"time": "75.331",
"value": "wong"
},
{
"operation": "Page search",
"time": "75.925",
"value": "wong"
},
{
"operation": "Highlighting",
"time": "82.932",
"value": "Terrorism in China_sentence_18"
},
{
"operation": "Highlighting",
"time": "90.255",
"value": "Terrorism in China_sentence_20"
},
{
"operation": "Highlighting deleted",
"time": "92.092",
"value": "Terrorism in China_sentence_20"
},
{
"operation": "Highlighting",
"time": "97.097",
"value": "Terrorism in China_sentence_19"
},
{
"operation": "finish",
"time": "111.31",
"value": "finish"
}
]
}
|
67025
|
Walt Kreinheder (September 2 1911 – October 22, 1970) was an Australian football player and his position was center and guard.
|
{
"evidence": [
"Walt Kreinheder_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:\nWalt Kreinheder (September 2 1911 – October 22, 1970) was an Australian football player and his position was center and guard.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Walt Kreinheder_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 682,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.251",
"value": "?search="
},
{
"operation": "search",
"time": "8.581",
"value": "Walt Kreinheder"
},
{
"operation": "Now on",
"time": "9.661",
"value": "Walt Kreinheder"
},
{
"operation": "Highlighting",
"time": "15.8",
"value": "Walt Kreinheder_cell_0_0_1"
},
{
"operation": "Highlighting",
"time": "15.809",
"value": "Walt Kreinheder_header_cell_0_0_0"
},
{
"operation": "Highlighting deleted",
"time": "31.054",
"value": "Walt Kreinheder_cell_0_0_1"
},
{
"operation": "Highlighting deleted",
"time": "31.064",
"value": "Walt Kreinheder_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "37.885",
"value": "Walt Kreinheder_sentence_0"
},
{
"operation": "finish",
"time": "64.094",
"value": "finish"
}
]
}
|
38635
|
George Seligman (born Abril 30, 1927 in New York)is an American Lie algebras mathematician and Professor Emeritus of Mathematics who served as chairman of the mathematics department of Yale University and a Fulbright Lecturer at the University of Münster.
|
{
"evidence": [
"George Seligman_sentence_0",
"George Seligman_sentence_4",
"George Seligman_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:\nGeorge Seligman (born Abril 30, 1927 in New York)is an American Lie algebras mathematician and Professor Emeritus of Mathematics who served as chairman of the mathematics department of Yale University and a Fulbright Lecturer at the University of Münster.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"George Seligman_sentence_0",
"George Seligman_sentence_4",
"George Seligman_sentence_3"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 683,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.523",
"value": "?search="
},
{
"operation": "search",
"time": "4.012",
"value": "George Seligman"
},
{
"operation": "Now on",
"time": "4.831",
"value": "George Seligman"
},
{
"operation": "Highlighting",
"time": "68.336",
"value": "George Seligman_sentence_0"
},
{
"operation": "Highlighting",
"time": "98.146",
"value": "George Seligman_sentence_4"
},
{
"operation": "Highlighting",
"time": "98.754",
"value": "George Seligman_sentence_3"
},
{
"operation": "finish",
"time": "384.935",
"value": "finish"
}
]
}
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.