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 |
|---|---|---|---|---|---|---|---|---|
15609
|
Koen Wauters started his career as a TV presenter in 1999, and with his brother Kris Wauters he presented all three seasons of the Flemish version of Pop Idol
|
{
"evidence": [
"Koen Wauters_sentence_9"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nKoen Wauters started his career as a TV presenter in 1999, and with his brother Kris Wauters he presented all three seasons of the Flemish version of Pop Idol\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Koen Wauters_sentence_9"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3384,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.279",
"value": "?search="
},
{
"operation": "search",
"time": "68.227",
"value": "Koen Wauters"
},
{
"operation": "Now on",
"time": "69.157",
"value": "Koen Wauters"
},
{
"operation": "Highlighting",
"time": "142.382",
"value": "Koen Wauters_sentence_9"
},
{
"operation": "finish",
"time": "167.911",
"value": "finish"
}
]
}
|
exfever_5102
|
Sabrina Annlynn Carpenter is an American actress whose breakthrough starring in the series Girl Meets World by an American pay television channel that serves as the flagship property of Disney Branded Television.
|
{
"evidence": [
"Sabrina Carpenter",
"Disney Channel"
],
"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:\nSabrina Annlynn Carpenter is an American actress whose breakthrough starring in the series Girl Meets World by an American pay television channel that serves as the flagship property of Disney Branded Television.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Sabrina Carpenter",
"Disney Channel"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3385,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8198
|
Rock music is popular music, with numerous subgenres, such as?an instrument that was amplified and used by?an American blues musician who was an innovator of electric blues sounds.
|
{
"evidence": [
"Rock music",
"Electric blues",
"T-Bone Walker"
],
"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:\nRock music is popular music, with numerous subgenres, such as?an instrument that was amplified and used by?an American blues musician who was an innovator of electric blues sounds.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Rock music",
"Electric blues",
"T-Bone Walker"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3386,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2878
|
Snoop Dogg became the most prominent rapper in America.
|
{
"evidence": [
"Snoop Dogg",
"Deep Cover (song)"
],
"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:\nSnoop Dogg became the most prominent rapper in America.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Snoop Dogg",
"Deep Cover (song)"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3387,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4817
|
Muhammad was an Arab religious and political leader and the founder of the second-largest religion, whose central text, the Quran, is considered by Muslims to be the word of God.
|
{
"evidence": [
"Muhammad",
"Islam"
],
"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:\nMuhammad was an Arab religious and political leader and the founder of the second-largest religion, whose central text, the Quran, is considered by Muslims to be the word of God.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Muhammad",
"Islam"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3388,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8055
|
Alan Alda is also nominated for Grammy Awards.
|
{
"evidence": [
"Bridge of Spies (film)",
"Alan Alda"
],
"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:\nAlan Alda is also nominated for Grammy Awards.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Bridge of Spies (film)",
"Alan Alda"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3389,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
35123
|
Resolution 434 was adopted in 1978 by the The United Nations Security Council, one of the six principal organs of the United Nations.
|
{
"evidence": [
"United Nations Security Council Resolution 434_sentence_0",
"United Nations Security Council_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:\nResolution 434 was adopted in 1978 by the The United Nations Security Council, one of the six principal organs of the United Nations.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"United Nations Security Council Resolution 434_sentence_0",
"United Nations Security Council_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3390,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.441",
"value": "?search="
},
{
"operation": "search",
"time": "12.71",
"value": "Resolution 434"
},
{
"operation": "Now on",
"time": "14.432",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "19.294",
"value": "United Nations Security Council Resolution 434"
},
{
"operation": "Now on",
"time": "20.371",
"value": "United Nations Security Council Resolution 434"
},
{
"operation": "Highlighting",
"time": "27.541",
"value": "United Nations Security Council Resolution 434_sentence_0"
},
{
"operation": "hyperlink",
"time": "56.165",
"value": "Security Council Resolution"
},
{
"operation": "Now on",
"time": "57.567",
"value": "List of United Nations resolutions concerning Israel"
},
{
"operation": "back-button-clicked",
"time": "67.335",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "67.789",
"value": "United Nations Security Council Resolution 434"
},
{
"operation": "hyperlink",
"time": "137.831",
"value": "Security Council"
},
{
"operation": "Now on",
"time": "138.985",
"value": "United Nations Security Council"
},
{
"operation": "Highlighting",
"time": "142.905",
"value": "United Nations Security Council_sentence_2"
},
{
"operation": "finish",
"time": "201.797",
"value": "finish"
}
]
}
|
exfever_4674
|
Zac Efron is an American actor who began acting professionally and also starred in the American fantasy action film directed by Burr Steers.
|
{
"evidence": [
"Zac Efron",
"17 Again (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:\nZac Efron is an American actor who began acting professionally and also starred in the American fantasy action film directed by Burr Steers.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Zac Efron",
"17 Again (film)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3391,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
13447
|
The squadron was activated at Shaw Air Force Base, South Carolina; then was awarded an Air Force Outstanding Unit Award (established in 1954) in 1962, for its support of reconnaissance operations during the Cuban Missile Crisis.
|
{
"evidence": [
"363rd Intelligence, Surveillance and Reconnaissance Wing_sentence_22",
"363rd Intelligence, Surveillance and Reconnaissance Wing_sentence_37",
"363rd Intelligence, Surveillance and Reconnaissance Wing_sentence_38",
"363rd Intelligence, Surveillance and Reconnaissance Wing_sentence_39",
"Air Force Outstanding Unit Award_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 squadron was activated at Shaw Air Force Base, South Carolina; then was awarded an Air Force Outstanding Unit Award (established in 1954) in 1962, for its support of reconnaissance operations during the Cuban Missile Crisis.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"363rd Intelligence, Surveillance and Reconnaissance Wing_sentence_22",
"363rd Intelligence, Surveillance and Reconnaissance Wing_sentence_37",
"363rd Intelligence, Surveillance and Reconnaissance Wing_sentence_38",
"363rd Intelligence, Surveillance and Reconnaissance Wing_sentence_39",
"Air Force Outstanding Unit Award_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3392,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.907",
"value": "?search="
},
{
"operation": "search",
"time": "28.729",
"value": "363rd Intelligence, Surveillance and Reconnaissance Wing"
},
{
"operation": "Now on",
"time": "29.669",
"value": "363rd Intelligence, Surveillance and Reconnaissance Wing"
},
{
"operation": "Page search",
"time": "35.124",
"value": "Shaw Air Force Base"
},
{
"operation": "Page search",
"time": "58.054",
"value": "Air Force Outstanding Unit Award"
},
{
"operation": "Page search",
"time": "145.475",
"value": "Shaw Air Force Base"
},
{
"operation": "Highlighting",
"time": "148.572",
"value": "363rd Intelligence, Surveillance and Reconnaissance Wing_sentence_22"
},
{
"operation": "Page search",
"time": "165.025",
"value": "Air Force Outstanding Unit Award"
},
{
"operation": "Highlighting",
"time": "176.013",
"value": "363rd Intelligence, Surveillance and Reconnaissance Wing_sentence_37"
},
{
"operation": "Highlighting",
"time": "196.965",
"value": "363rd Intelligence, Surveillance and Reconnaissance Wing_sentence_39"
},
{
"operation": "Highlighting deleted",
"time": "200.402",
"value": "363rd Intelligence, Surveillance and Reconnaissance Wing_sentence_39"
},
{
"operation": "Highlighting",
"time": "200.82",
"value": "363rd Intelligence, Surveillance and Reconnaissance Wing_sentence_38"
},
{
"operation": "Highlighting",
"time": "201.37",
"value": "363rd Intelligence, Surveillance and Reconnaissance Wing_sentence_39"
},
{
"operation": "hyperlink",
"time": "221.497",
"value": "Air Force Outstanding Unit Award"
},
{
"operation": "Now on",
"time": "222.216",
"value": "Air Force Outstanding Unit Award"
},
{
"operation": "Highlighting",
"time": "225.645",
"value": "Air Force Outstanding Unit Award_sentence_1"
},
{
"operation": "back-button-clicked",
"time": "231.288",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "231.393",
"value": "363rd Intelligence, Surveillance and Reconnaissance Wing"
},
{
"operation": "finish",
"time": "241.972",
"value": "finish"
}
]
}
|
91278
|
Lord Howe Island is an important transit and refueling point, it is located 600 km (373 mi) to the west is the Australian mainland whose population is about 24.5 million and some countries are close to the Island.
|
{
"evidence": [
"Lord Howe Island_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:\nLord Howe Island is an important transit and refueling point, it is located 600 km (373 mi) to the west is the Australian mainland whose population is about 24.5 million and some countries are close to the Island.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Lord Howe Island_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3393,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "4.237",
"value": "undefined"
},
{
"operation": "search",
"time": "24.871",
"value": "Lord Howe Island"
},
{
"operation": "Now on",
"time": "27.357",
"value": "Lord Howe Island"
},
{
"operation": "Highlighting",
"time": "47.529",
"value": "Lord Howe Island_sentence_1"
},
{
"operation": "Page search",
"time": "87.818",
"value": "fuel"
},
{
"operation": "Page search",
"time": "89.112",
"value": "fuel"
},
{
"operation": "Page search",
"time": "89.322",
"value": "fuel"
},
{
"operation": "Page search",
"time": "89.489",
"value": "fuel"
},
{
"operation": "Page search",
"time": "89.625",
"value": "fuel"
},
{
"operation": "Page search",
"time": "147.764",
"value": "transit"
},
{
"operation": "Page search",
"time": "148.176",
"value": "transit"
},
{
"operation": "Page search",
"time": "148.344",
"value": "transit"
},
{
"operation": "Page search",
"time": "148.512",
"value": "transit"
},
{
"operation": "Page search",
"time": "148.643",
"value": "transit"
},
{
"operation": "Page search",
"time": "148.76",
"value": "transit"
},
{
"operation": "Highlighting",
"time": "250.13",
"value": "Lord Howe Island_sentence_10"
},
{
"operation": "Highlighting",
"time": "255.251",
"value": "Lord Howe Island_sentence_11"
},
{
"operation": "Highlighting",
"time": "266.721",
"value": "Lord Howe Island_sentence_12"
},
{
"operation": "Highlighting deleted",
"time": "399.556",
"value": "Lord Howe Island_sentence_10"
},
{
"operation": "Highlighting deleted",
"time": "400.523",
"value": "Lord Howe Island_sentence_12"
},
{
"operation": "Highlighting deleted",
"time": "401.287",
"value": "Lord Howe Island_sentence_11"
},
{
"operation": "finish",
"time": "436.862",
"value": "finish"
}
]
}
|
80681
|
The Order of the Golden Fleece is closely associated with the Holy Roman Empire, but their is a splinter group that adheres to the newer statues of the order.
|
{
"evidence": [
"Order of the Golden Fleece_sentence_43",
"Order of the Golden Fleece_sentence_44",
"Order of the Golden Fleece_sentence_45"
],
"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 Order of the Golden Fleece is closely associated with the Holy Roman Empire, but their is a splinter group that adheres to the newer statues of the order.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Order of the Golden Fleece_sentence_43",
"Order of the Golden Fleece_sentence_44",
"Order of the Golden Fleece_sentence_45"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3394,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.38",
"value": "?search="
},
{
"operation": "search",
"time": "12.191",
"value": "Order of the Golden Fleece"
},
{
"operation": "Now on",
"time": "13.194",
"value": "Order of the Golden Fleece"
},
{
"operation": "Highlighting",
"time": "61.308",
"value": "Order of the Golden Fleece_sentence_43"
},
{
"operation": "Highlighting",
"time": "118.281",
"value": "Order of the Golden Fleece_sentence_44"
},
{
"operation": "Highlighting",
"time": "120.802",
"value": "Order of the Golden Fleece_sentence_45"
},
{
"operation": "finish",
"time": "152.229",
"value": "finish"
}
]
}
|
91758
|
Ray Bolger demanded that the producers expand his role, then several rewrites later, de Mille herself was fired and replaced by Burt Shevelove (an American musical theater playwright); but only a few days after that, Shevelove exited.
|
{
"evidence": [
"Agnes de Mille_sentence_0",
"Agnes de Mille_sentence_1",
"Burt Shevelove_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:\nRay Bolger demanded that the producers expand his role, then several rewrites later, de Mille herself was fired and replaced by Burt Shevelove (an American musical theater playwright); but only a few days after that, Shevelove exited.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Agnes de Mille_sentence_0",
"Agnes de Mille_sentence_1",
"Burt Shevelove_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3395,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.714",
"value": "undefined"
},
{
"operation": "search",
"time": "9.606",
"value": "Roy Bolger"
},
{
"operation": "Now on",
"time": "11.962",
"value": "undefined"
},
{
"operation": "search",
"time": "19.238",
"value": "Ray Bolger"
},
{
"operation": "Now on",
"time": "20.157",
"value": "Ray Bolger"
},
{
"operation": "search",
"time": "122.15",
"value": "Burt Shevelove"
},
{
"operation": "Now on",
"time": "122.753",
"value": "Burt Shevelove"
},
{
"operation": "Highlighting",
"time": "125.15",
"value": "Burt Shevelove_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "126.576",
"value": "Burt Shevelove_sentence_0"
},
{
"operation": "search",
"time": "150.082",
"value": "De Mille"
},
{
"operation": "Now on",
"time": "151.024",
"value": "DeMille"
},
{
"operation": "hyperlink",
"time": "169.501",
"value": "Katherine DeMille"
},
{
"operation": "Now on",
"time": "170.276",
"value": "Katherine DeMille"
},
{
"operation": "search",
"time": "270.556",
"value": "Ray Bolger"
},
{
"operation": "Now on",
"time": "271.189",
"value": "Ray Bolger"
},
{
"operation": "hyperlink",
"time": "280.807",
"value": "Scarecrow"
},
{
"operation": "Now on",
"time": "281.592",
"value": "Scarecrow (Oz)"
},
{
"operation": "search",
"time": "340.304",
"value": "Ray Bolger"
},
{
"operation": "Now on",
"time": "341.429",
"value": "Ray Bolger"
},
{
"operation": "hyperlink",
"time": "354.526",
"value": "Babes in Toyland"
},
{
"operation": "Now on",
"time": "355.271",
"value": "Babes in Toyland (1961 film)"
},
{
"operation": "back-button-clicked",
"time": "386.674",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "386.782",
"value": "Ray Bolger"
},
{
"operation": "hyperlink",
"time": "409.56",
"value": "The Great Ziegfeld"
},
{
"operation": "Now on",
"time": "410.366",
"value": "The Great Ziegfeld"
},
{
"operation": "back-button-clicked",
"time": "456.892",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "457.008",
"value": "Ray Bolger"
},
{
"operation": "hyperlink",
"time": "461.634",
"value": "Sweethearts"
},
{
"operation": "Now on",
"time": "462.422",
"value": "Sweethearts (1938 film)"
},
{
"operation": "hyperlink",
"time": "476.331",
"value": "Ray Bolger"
},
{
"operation": "Now on",
"time": "476.755",
"value": "Ray Bolger"
},
{
"operation": "hyperlink",
"time": "503.061",
"value": "The Harvey Girls"
},
{
"operation": "Now on",
"time": "503.798",
"value": "The Harvey Girls"
},
{
"operation": "back-button-clicked",
"time": "528.554",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "528.67",
"value": "Ray Bolger"
},
{
"operation": "search",
"time": "591.202",
"value": "Burt Shevelove"
},
{
"operation": "Now on",
"time": "591.767",
"value": "Burt Shevelove"
},
{
"operation": "search",
"time": "622.358",
"value": "Contains...de Mille"
},
{
"operation": "Now on",
"time": "625.116",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "634.41",
"value": "Brigadoon"
},
{
"operation": "Now on",
"time": "635.223",
"value": "Brigadoon"
},
{
"operation": "back-button-clicked",
"time": "734.746",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "735.381",
"value": "undefined"
},
{
"operation": "search",
"time": "745.672",
"value": "Agnes de Mille"
},
{
"operation": "Now on",
"time": "746.701",
"value": "Agnes de Mille"
},
{
"operation": "Highlighting",
"time": "748.409",
"value": "Agnes de Mille_sentence_0"
},
{
"operation": "Highlighting",
"time": "816.728",
"value": "Agnes de Mille_sentence_1"
},
{
"operation": "search",
"time": "829.569",
"value": "Burt Shevelove"
},
{
"operation": "Now on",
"time": "830.149",
"value": "Burt Shevelove"
},
{
"operation": "Highlighting",
"time": "851.901",
"value": "Burt Shevelove_sentence_0"
},
{
"operation": "finish",
"time": "863.66",
"value": "finish"
}
]
}
|
74088
|
P2RX7 (a protein that in humans is encoded by the P2RX7 gene) subunits can form homomeric receptors and the receptor is a ligand-gated cation channel which requires lower levels of ATP.
|
{
"evidence": [
"P2RX7_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:\nP2RX7 (a protein that in humans is encoded by the P2RX7 gene) subunits can form homomeric receptors and the receptor is a ligand-gated cation channel which requires lower levels of ATP.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"P2RX7_sentence_7"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3396,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.414",
"value": "?search="
},
{
"operation": "search",
"time": "10.89",
"value": "P2RX7"
},
{
"operation": "Now on",
"time": "12.641",
"value": "P2RX7"
},
{
"operation": "Highlighting",
"time": "14.889",
"value": "P2RX7_sentence_0"
},
{
"operation": "Highlighting",
"time": "38.946",
"value": "P2RX7_sentence_5"
},
{
"operation": "Highlighting",
"time": "43.564",
"value": "P2RX7_sentence_6"
},
{
"operation": "Highlighting",
"time": "52.004",
"value": "P2RX7_sentence_7"
},
{
"operation": "Highlighting deleted",
"time": "53.186",
"value": "P2RX7_sentence_6"
},
{
"operation": "Highlighting deleted",
"time": "53.98",
"value": "P2RX7_sentence_5"
},
{
"operation": "Highlighting deleted",
"time": "58.635",
"value": "P2RX7_sentence_0"
},
{
"operation": "finish",
"time": "68.461",
"value": "finish"
}
]
}
|
exfever_4313
|
Tom Felton is an English actor best known for playing Draco Malfoy in the film adaptations of a series of seven novels set in an imaginary universe, often but not always without any locations, events, or people from the real world.
|
{
"evidence": [
"Tom Felton",
"Harry Potter",
"Fantasy literature"
],
"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 Felton is an English actor best known for playing Draco Malfoy in the film adaptations of a series of seven novels set in an imaginary universe, often but not always without any locations, events, or people from the real world.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Tom Felton",
"Harry Potter",
"Fantasy literature"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3397,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
57470
|
Eumelea ludovicata has a yellowish white round-shaped head and hairs that are short, erect, and thick.
|
{
"evidence": [
"Eumelea ludovicata_sentence_6",
"Eumelea ludovicata_sentence_7"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nEumelea ludovicata has a yellowish white round-shaped head and hairs that are short, erect, and thick.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Eumelea ludovicata_sentence_6",
"Eumelea ludovicata_sentence_7"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3398,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.502",
"value": "?search="
},
{
"operation": "search",
"time": "92.782",
"value": "Eumelea ludovicata"
},
{
"operation": "Now on",
"time": "93.732",
"value": "Eumelea ludovicata"
},
{
"operation": "Highlighting",
"time": "112.735",
"value": "Eumelea ludovicata_sentence_6"
},
{
"operation": "Highlighting",
"time": "117.887",
"value": "Eumelea ludovicata_sentence_7"
},
{
"operation": "Highlighting deleted",
"time": "160.903",
"value": "Eumelea ludovicata_sentence_7"
},
{
"operation": "Highlighting",
"time": "202.423",
"value": "Eumelea ludovicata_sentence_7"
},
{
"operation": "finish",
"time": "214.431",
"value": "finish"
}
]
}
|
exfever_5533
|
Wilhelmina Slater is a fictional character played by Vanessa Williams, for whom she received a supporting actress nomination at the ceremony that honors the best in the U.S. block of broadcast programming taking place during the middle of the evening for a television show.
|
{
"evidence": [
"Wilhelmina Slater",
"59th Primetime Emmy Awards",
"Prime time"
],
"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:\nWilhelmina Slater is a fictional character played by Vanessa Williams, for whom she received a supporting actress nomination at the ceremony that honors the best in the U.S. block of broadcast programming taking place during the middle of the evening for a television show.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Wilhelmina Slater",
"59th Primetime Emmy Awards",
"Prime time"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3399,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
36240
|
In 2000, Faith became a member of No Secrets and their first single, "That's What Girls Do", was a hit and was also used in The Powerpuff Girls Movie, and later that year, was used in The Hot Chick and in 2002 after much success with the group, they toured alongside Aaron Carter on his "Oh Aaron tour" as a part of a Jive Records collaboration.
|
{
"evidence": [
"Angel Faith_sentence_10",
"Angel Faith_sentence_13",
"Angel Faith_sentence_14"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nIn 2000, Faith became a member of No Secrets and their first single, \"That's What Girls Do\", was a hit and was also used in The Powerpuff Girls Movie, and later that year, was used in The Hot Chick and in 2002 after much success with the group, they toured alongside Aaron Carter on his \"Oh Aaron tour\" as a part of a Jive Records collaboration.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Angel Faith_sentence_10",
"Angel Faith_sentence_13",
"Angel Faith_sentence_14"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3400,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.654",
"value": "?search="
},
{
"operation": "search",
"time": "23.7",
"value": "No Secrets"
},
{
"operation": "Now on",
"time": "24.465",
"value": "No Secrets"
},
{
"operation": "hyperlink",
"time": "48.581",
"value": "No Secrets (No Secrets album)"
},
{
"operation": "Now on",
"time": "49.12",
"value": "No Secrets (No Secrets album)"
},
{
"operation": "Page search",
"time": "61.67",
"value": "faith"
},
{
"operation": "Page search",
"time": "96.261",
"value": "girls"
},
{
"operation": "search",
"time": "137.508",
"value": "That's What Girls Do"
},
{
"operation": "Now on",
"time": "138.402",
"value": "No Secrets (No Secrets album)"
},
{
"operation": "Page search",
"time": "182.955",
"value": "single"
},
{
"operation": "back-button-clicked",
"time": "200.659",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "200.891",
"value": "No Secrets (No Secrets album)"
},
{
"operation": "search",
"time": "212.172",
"value": "That's What Girls Do"
},
{
"operation": "Now on",
"time": "212.713",
"value": "That's What Girls Do"
},
{
"operation": "hyperlink",
"time": "217.357",
"value": "That's What Girls Do"
},
{
"operation": "Now on",
"time": "218.015",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "222.192",
"value": "No Secrets (No Secrets album)"
},
{
"operation": "Now on",
"time": "222.47",
"value": "No Secrets (No Secrets album)"
},
{
"operation": "Highlighting",
"time": "257.157",
"value": "No Secrets (No Secrets album)_sentence_2"
},
{
"operation": "Highlighting",
"time": "257.158",
"value": "No Secrets (No Secrets album)_title"
},
{
"operation": "search",
"time": "306.993",
"value": "oh aaron tour"
},
{
"operation": "Now on",
"time": "308.516",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "319.224",
"value": "Angel Faith"
},
{
"operation": "Now on",
"time": "319.831",
"value": "Angel Faith"
},
{
"operation": "Highlighting",
"time": "340.588",
"value": "Angel Faith_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "360.045",
"value": "Angel Faith_sentence_1"
},
{
"operation": "Highlighting",
"time": "367.009",
"value": "Angel Faith_sentence_10"
},
{
"operation": "Highlighting",
"time": "373.634",
"value": "Angel Faith_sentence_13"
},
{
"operation": "Highlighting",
"time": "414.19",
"value": "Angel Faith_sentence_14"
},
{
"operation": "finish",
"time": "533.488",
"value": "finish"
}
]
}
|
exfever_2507
|
Muhammad was the founder of the Abrahamic monotheistic religion whose Quran is considered to be the word of the supreme being, creator, and principal object of faith.
|
{
"evidence": [
"Muhammad",
"Islam",
"God"
],
"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:\nMuhammad was the founder of the Abrahamic monotheistic religion whose Quran is considered to be the word of the supreme being, creator, and principal object of faith.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Muhammad",
"Islam",
"God"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3401,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_639
|
Romeo and Juliet is William Shakespeare's second tragedy.
|
{
"evidence": [
"Judith Quiney",
"Hamnet Shakespeare",
"Hamlet"
],
"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:\nRomeo and Juliet is William Shakespeare's second tragedy.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Judith Quiney",
"Hamnet Shakespeare",
"Hamlet"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3402,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
60450
|
Catherine Karlsdotter was the second wife of Charles VIII / I who was crowned Queen of Sweden in Uppsala Cathedral on 2 July 1448.
|
{
"evidence": [
"Catherine Karlsdotter_sentence_0",
"Catherine Karlsdotter_sentence_6"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nCatherine Karlsdotter was the second wife of Charles VIII / I who was crowned Queen of Sweden in Uppsala Cathedral on 2 July 1448.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Catherine Karlsdotter_sentence_0",
"Catherine Karlsdotter_sentence_6"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3403,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.698",
"value": "?search="
},
{
"operation": "search",
"time": "10.444",
"value": "Catherine Karlsdotter"
},
{
"operation": "Now on",
"time": "11.145",
"value": "Catherine Karlsdotter"
},
{
"operation": "Highlighting",
"time": "53.524",
"value": "Catherine Karlsdotter_sentence_0"
},
{
"operation": "Page search",
"time": "79.819",
"value": "crown"
},
{
"operation": "Highlighting",
"time": "86.006",
"value": "Catherine Karlsdotter_sentence_6"
},
{
"operation": "finish",
"time": "142.802",
"value": "finish"
}
]
}
|
exfever_2705
|
The 2014 WTA Finals were a women's tennis tournament held in the indoor arena located in Kallang.
|
{
"evidence": [
"2014 WTA Finals",
"Singapore Indoor Stadium"
],
"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 2014 WTA Finals were a women's tennis tournament held in the indoor arena located in Kallang.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"2014 WTA Finals",
"Singapore Indoor Stadium"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3404,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7390
|
The existence of God is a subject of debate in the philosophy of religion and mass culture.?
|
{
"evidence": [
"Existence of God",
"Popular culture"
],
"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 existence of God is a subject of debate in the philosophy of religion and mass culture.?\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Existence of God",
"Popular culture"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3405,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
1922
|
Erna Rebecca Takazawa was born December 21, 1988, in Japan, raised in Samoa and graduated with First Class Honors, an optometry degree from the University of Auckland, in 2011 then licensed, in 2012 she then immediately began working at Samoa Vision Centre and in conjunction with the National Health Service and SENSE a support program for special needs children wherein 2013 she screened over 200 athletes with the Special Olympics which is the world's largest sports organization for children and adults with intellectual and physical disabilities.
|
{
"evidence": [
"Erna Takazawa_sentence_2",
"Erna Takazawa_sentence_3",
"Erna Takazawa_sentence_4",
"Erna Takazawa_sentence_5",
"Erna Takazawa_sentence_7",
"Special Olympics_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:\nErna Rebecca Takazawa was born December 21, 1988, in Japan, raised in Samoa and graduated with First Class Honors, an optometry degree from the University of Auckland, in 2011 then licensed, in 2012 she then immediately began working at Samoa Vision Centre and in conjunction with the National Health Service and SENSE a support program for special needs children wherein 2013 she screened over 200 athletes with the Special Olympics which is the world's largest sports organization for children and adults with intellectual and physical disabilities.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Erna Takazawa_sentence_2",
"Erna Takazawa_sentence_3",
"Erna Takazawa_sentence_4",
"Erna Takazawa_sentence_5",
"Erna Takazawa_sentence_7",
"Special Olympics_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3406,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "search",
"time": "22.848",
"value": "Erna Rebecca Takazawa"
},
{
"operation": "hyperlink",
"time": "32.94",
"value": "Erna Takazawa"
},
{
"operation": "Highlighting",
"time": "69.23",
"value": "Erna Takazawa_cell_0_1_1"
},
{
"operation": "Highlighting",
"time": "69.254",
"value": "Erna Takazawa_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "69.26",
"value": "Erna Takazawa_header_cell_0_0_0"
},
{
"operation": "Highlighting deleted",
"time": "82.143",
"value": "Erna Takazawa_cell_0_1_1"
},
{
"operation": "Highlighting deleted",
"time": "82.173",
"value": "Erna Takazawa_header_cell_0_1_0"
},
{
"operation": "Highlighting deleted",
"time": "82.181",
"value": "Erna Takazawa_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "82.997",
"value": "Erna Takazawa_sentence_2"
},
{
"operation": "Highlighting",
"time": "86.558",
"value": "Erna Takazawa_sentence_3"
},
{
"operation": "Highlighting",
"time": "96.097",
"value": "Erna Takazawa_sentence_4"
},
{
"operation": "Highlighting",
"time": "101.38",
"value": "Erna Takazawa_sentence_5"
},
{
"operation": "Highlighting",
"time": "117.3",
"value": "Erna Takazawa_sentence_7"
},
{
"operation": "hyperlink",
"time": "125.334",
"value": "Special Olympics"
},
{
"operation": "Highlighting",
"time": "128.63",
"value": "Special Olympics_sentence_1"
},
{
"operation": "finish",
"time": "135.983",
"value": "finish"
}
]
}
|
exfever_4062
|
The Beatles were an English rock band formed in Liverpool in 1960 that played an integral role in developing the anti-establishment cultural phenomenon that developed throughout much of the Western world between the mid-1960s and the mid-1970s.
|
{
"evidence": [
"The Beatles",
"Counterculture of the 1960s"
],
"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 Beatles were an English rock band formed in Liverpool in 1960 that played an integral role in developing the anti-establishment cultural phenomenon that developed throughout much of the Western world between the mid-1960s and the mid-1970s.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Beatles",
"Counterculture of the 1960s"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3407,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_427
|
Wentworth is an Australian television drama series that initially focuses on the early days in prison of Bea Smith, a character portrayed by a New Zealand stage and screen actress known for her role as Rachel Greene.
|
{
"evidence": [
"Wentworth (TV series)",
"Danielle Cormack"
],
"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:\nWentworth is an Australian television drama series that initially focuses on the early days in prison of Bea Smith, a character portrayed by a New Zealand stage and screen actress known for her role as Rachel Greene.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Wentworth (TV series)",
"Danielle Cormack"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3408,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
47714
|
Samarth did his schooling at the Bishop Cotton Boys' School and B'Com at SBM Jain College in Bangalore, while training at the Karnataka Institute of Cricket.
|
{
"evidence": [
"Ravikumar Samarth_sentence_4",
"Ravikumar Samarth_sentence_5"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nSamarth did his schooling at the Bishop Cotton Boys' School and B'Com at SBM Jain College in Bangalore, while training at the Karnataka Institute of Cricket.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Ravikumar Samarth_sentence_4",
"Ravikumar Samarth_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3409,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.703",
"value": "?search="
},
{
"operation": "search",
"time": "12.2",
"value": "Contains...Bishop Cotton Boys' School"
},
{
"operation": "Now on",
"time": "14.035",
"value": "undefined"
},
{
"operation": "search",
"time": "53.1",
"value": "Ravikumar Samarth"
},
{
"operation": "Now on",
"time": "53.775",
"value": "Ravikumar Samarth"
},
{
"operation": "Page search",
"time": "61.254",
"value": "Bishop Cotton Boys' School"
},
{
"operation": "Page search",
"time": "61.424",
"value": "Bishop Cotton Boys' School"
},
{
"operation": "Highlighting",
"time": "62.989",
"value": "Ravikumar Samarth_sentence_4"
},
{
"operation": "Highlighting",
"time": "69.734",
"value": "Ravikumar Samarth_sentence_5"
},
{
"operation": "finish",
"time": "491.855",
"value": "finish"
}
]
}
|
44392
|
The Palazzo Bonacolsi is a 13th-century Gothic-style aristocratic brick palace built in the late thirteenth century by the Pinamonte Baldassare.
|
{
"evidence": [
"Palazzo Bonacolsi_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:\nThe Palazzo Bonacolsi is a 13th-century Gothic-style aristocratic brick palace built in the late thirteenth century by the Pinamonte Baldassare.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Palazzo Bonacolsi_sentence_7"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3410,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.922",
"value": "?search="
},
{
"operation": "search",
"time": "10.306",
"value": "Contains...The Palazzo Bonacolsi"
},
{
"operation": "Now on",
"time": "11.239",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "15.801",
"value": "Palazzo Bonacolsi"
},
{
"operation": "Now on",
"time": "16.451",
"value": "Palazzo Bonacolsi"
},
{
"operation": "Highlighting",
"time": "23.402",
"value": "Palazzo Bonacolsi_sentence_0"
},
{
"operation": "Highlighting",
"time": "41.295",
"value": "Palazzo Bonacolsi_sentence_5"
},
{
"operation": "Highlighting",
"time": "47.494",
"value": "Palazzo Bonacolsi_sentence_7"
},
{
"operation": "Page search",
"time": "80.137",
"value": "Baldassare"
},
{
"operation": "Highlighting deleted",
"time": "123.322",
"value": "Palazzo Bonacolsi_sentence_5"
},
{
"operation": "Highlighting deleted",
"time": "124.578",
"value": "Palazzo Bonacolsi_sentence_0"
},
{
"operation": "finish",
"time": "135.601",
"value": "finish"
}
]
}
|
exfever_6774
|
Mark Alan Ruffalo went on to star in a 2010 American neo-noir psychological horror film based on a novel published by Harper Collins.
|
{
"evidence": [
"Mark Ruffalo",
"Shutter Island (film)"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nMark Alan Ruffalo went on to star in a 2010 American neo-noir psychological horror film based on a novel published by Harper Collins.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Mark Ruffalo",
"Shutter Island (film)"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3411,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_264
|
John Paul Cusack starred in a 1998 comedy film starring an American actor, filmmaker, and musician.
|
{
"evidence": [
"John Cusack",
"Tapeheads",
"Tim Robbins"
],
"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:\nJohn Paul Cusack starred in a 1998 comedy film starring an American actor, filmmaker, and musician.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"John Cusack",
"Tapeheads",
"Tim Robbins"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3412,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
36661
|
In the WA version of clout archery, the distance for the recurve bows event for men is farther by 20m (45 yd) compared to the women's.
|
{
"evidence": [
"Clout archery_sentence_24"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nIn the WA version of clout archery, the distance for the recurve bows event for men is farther by 20m (45 yd) compared to the women's.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Clout archery_sentence_24"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3413,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.747",
"value": "?search="
},
{
"operation": "search",
"time": "16.439",
"value": "wa clout archery"
},
{
"operation": "Now on",
"time": "17.572",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "20.774",
"value": "Clout archery"
},
{
"operation": "Now on",
"time": "21.315",
"value": "Clout archery"
},
{
"operation": "hyperlink",
"time": "37.606",
"value": "3 WA clout Clout archery_section_2"
},
{
"operation": "Highlighting",
"time": "68.622",
"value": "Clout archery_sentence_24"
},
{
"operation": "finish",
"time": "87.365",
"value": "finish"
}
]
}
|
86071
|
Memel often sought city rights, but was opposed by Heydekrug (Šilutė) (which was a gathering place for peasants from nearby Samogitia and Curonian and Prussian fishermen from Rusnė, Karklė ) and Tilsit in 1721 and 1725.
|
{
"evidence": [
"Šilutė_sentence_9"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nMemel often sought city rights, but was opposed by Heydekrug (Šilutė) (which was a gathering place for peasants from nearby Samogitia and Curonian and Prussian fishermen from Rusnė, Karklė ) and Tilsit in 1721 and 1725.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Šilutė_sentence_9"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3414,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.78",
"value": "?search="
},
{
"operation": "search",
"time": "11.43",
"value": "Memel"
},
{
"operation": "Now on",
"time": "12.116",
"value": "Memel"
},
{
"operation": "hyperlink",
"time": "52.186",
"value": "Memel Territory"
},
{
"operation": "Now on",
"time": "53.398",
"value": "Klaipėda Region"
},
{
"operation": "search",
"time": "87.384",
"value": "Contains...Memel often sought city right"
},
{
"operation": "Now on",
"time": "89.309",
"value": "undefined"
},
{
"operation": "search",
"time": "104.347",
"value": "Memel often sought city rights"
},
{
"operation": "Now on",
"time": "105.267",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "113.126",
"value": "Memel League of Nations_section_14"
},
{
"operation": "Now on",
"time": "116.058",
"value": "League of Nations#Memel League of Nations section 14"
},
{
"operation": "search",
"time": "181.904",
"value": "memel was opposed by Heydekrug"
},
{
"operation": "Now on",
"time": "183.159",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "192.579",
"value": "Šilutė"
},
{
"operation": "Now on",
"time": "193.25",
"value": "Šilutė"
},
{
"operation": "Page search",
"time": "208.597",
"value": "city rights"
},
{
"operation": "Highlighting",
"time": "214.294",
"value": "Šilutė_sentence_9"
},
{
"operation": "finish",
"time": "268.996",
"value": "finish"
}
]
}
|
40359
|
The Australian Journal of Linguistics was established in 1981 as a quarterly publication by the Australian Linguistic Society.
|
{
"evidence": [
"Australian Journal of Linguistics_sentence_0",
"Australian Journal of Linguistics_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 Australian Journal of Linguistics was established in 1981 as a quarterly publication by the Australian Linguistic Society.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Australian Journal of Linguistics_sentence_0",
"Australian Journal of Linguistics_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3415,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.07",
"value": "?search="
},
{
"operation": "search",
"time": "6.462",
"value": "Australian Journal of Linguistics"
},
{
"operation": "Now on",
"time": "8.056",
"value": "Australian Journal of Linguistics"
},
{
"operation": "Highlighting",
"time": "11.154",
"value": "Australian Journal of Linguistics_sentence_0"
},
{
"operation": "Page search",
"time": "19.066",
"value": "quart"
},
{
"operation": "Highlighting deleted",
"time": "22.332",
"value": "Australian Journal of Linguistics_sentence_0"
},
{
"operation": "Highlighting",
"time": "22.876",
"value": "Australian Journal of Linguistics_sentence_0"
},
{
"operation": "Highlighting",
"time": "31.42",
"value": "Australian Journal of Linguistics_sentence_1"
},
{
"operation": "finish",
"time": "187.478",
"value": "finish"
}
]
}
|
exfever_1497
|
The Smurfs was the first film written by Max Botkin.
|
{
"evidence": [
"The Smurfs (film)",
"Raja Gosnell",
"Show Dogs"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Smurfs was the first film written by Max Botkin.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Smurfs (film)",
"Raja Gosnell",
"Show Dogs"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3416,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
37597
|
Bell Canyon Park (large open-space regional park) has trails along riparian Bell Creek and upstream beside the Bell Canyon community and the trails have junctions to connect with El Escorpión Park and Upper Las Virgenes Canyon Open Space Preserve and their miles of trail networks.
|
{
"evidence": [
"Bell Canyon Park_sentence_4",
"Bell Canyon Park_sentence_3",
"Bell Canyon Park_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:\nBell Canyon Park (large open-space regional park) has trails along riparian Bell Creek and upstream beside the Bell Canyon community and the trails have junctions to connect with El Escorpión Park and Upper Las Virgenes Canyon Open Space Preserve and their miles of trail networks.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Bell Canyon Park_sentence_4",
"Bell Canyon Park_sentence_3",
"Bell Canyon Park_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3417,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.607",
"value": "undefined"
},
{
"operation": "search",
"time": "18.24",
"value": "Bell Canyon Park ("
},
{
"operation": "Now on",
"time": "19.708",
"value": "undefined"
},
{
"operation": "search",
"time": "28.187",
"value": "Bell Canyon Park"
},
{
"operation": "Now on",
"time": "29.478",
"value": "Bell Canyon Park"
},
{
"operation": "Highlighting",
"time": "74.343",
"value": "Bell Canyon Park_sentence_4"
},
{
"operation": "Highlighting",
"time": "75.667",
"value": "Bell Canyon Park_sentence_3"
},
{
"operation": "Highlighting",
"time": "80.849",
"value": "Bell Canyon Park_sentence_0"
},
{
"operation": "Highlighting",
"time": "100.995",
"value": "Bell Canyon Park_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "171.707",
"value": "Bell Canyon Park_sentence_1"
},
{
"operation": "finish",
"time": "185.787",
"value": "finish"
}
]
}
|
exfever_8515
|
Soundgarden were an American rock band with a drummer for Pearl Jam and side projects such as a rock supergroup that formed in Seattle, Washington in 1995.
|
{
"evidence": [
"Soundgarden",
"Matt Cameron",
"Hater (band)"
],
"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:\nSoundgarden were an American rock band with a drummer for Pearl Jam and side projects such as a rock supergroup that formed in Seattle, Washington in 1995.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Soundgarden",
"Matt Cameron",
"Hater (band)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3418,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7216
|
Lindsay Dee Lohan is an actress who appeared in a 2004 American teen musical film with a genre in which songs by the characters are interwoven into the narrative.
|
{
"evidence": [
"Lindsay Lohan",
"Confessions of a Teenage Drama Queen",
"Musical 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:\nLindsay Dee Lohan is an actress who appeared in a 2004 American teen musical film with a genre in which songs by the characters are interwoven into the narrative.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Lindsay Lohan",
"Confessions of a Teenage Drama Queen",
"Musical film"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3419,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8686
|
One World Trade Center is the new skyscraper in the United Kingdom that has the same name as the North Tower of the large complex in the Financial District of Lower Manhattan designed by Minoru Yamasaki.
|
{
"evidence": [
"One World Trade Center",
"World Trade Center (1973–2001)"
],
"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:\nOne World Trade Center is the new skyscraper in the United Kingdom that has the same name as the North Tower of the large complex in the Financial District of Lower Manhattan designed by Minoru Yamasaki.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"One World Trade Center",
"World Trade Center (1973–2001)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3420,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
21314
|
Christopher Lewknor, who by the early 1630s was recorder of Chichester, was also elected MP for Chichester in the Short Parliament in 1640, and sat in the House of Commons variously between 1628 and 1641.
|
{
"evidence": [
"Christopher Lewknor_sentence_15",
"Christopher Lewknor_sentence_16",
"Christopher Lewknor_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:\nChristopher Lewknor, who by the early 1630s was recorder of Chichester, was also elected MP for Chichester in the Short Parliament in 1640, and sat in the House of Commons variously between 1628 and 1641.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Christopher Lewknor_sentence_15",
"Christopher Lewknor_sentence_16",
"Christopher Lewknor_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3421,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.999",
"value": "?search="
},
{
"operation": "search",
"time": "18.753",
"value": "Christopher Lewknor"
},
{
"operation": "Now on",
"time": "19.415",
"value": "Christopher Lewknor"
},
{
"operation": "Page search",
"time": "51.956",
"value": "recorder"
},
{
"operation": "Highlighting",
"time": "56.233",
"value": "Christopher Lewknor_sentence_15"
},
{
"operation": "Highlighting",
"time": "67.204",
"value": "Christopher Lewknor_sentence_16"
},
{
"operation": "Highlighting",
"time": "92.287",
"value": "Christopher Lewknor_sentence_0"
},
{
"operation": "finish",
"time": "128.501",
"value": "finish"
}
]
}
|
exfever_5806
|
Turkey is a transcontinental country located primarily on a large peninsula in Western Europe and the westernmost protrusion of the Asian continent that constitutes the major part of modern-day Turkey.
|
{
"evidence": [
"Turkey",
"Anatolia"
],
"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:\nTurkey is a transcontinental country located primarily on a large peninsula in Western Europe and the westernmost protrusion of the Asian continent that constitutes the major part of modern-day Turkey.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Turkey",
"Anatolia"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3422,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2534
|
The Offspring is an American rock band from a city in northern Orange County, Australia.
|
{
"evidence": [
"The Offspring",
"Garden Grove, California"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Offspring is an American rock band from a city in northern Orange County, Australia.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Offspring",
"Garden Grove, California"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3423,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8259
|
"Champion" is a 1949 American drama film noir sports film directed by Mark Robson and starring an American actor whose other early films included a 1945 film noir directed by Jacques Tourneur.
|
{
"evidence": [
"Champion (1949 film)",
"Kirk Douglas",
"Out of the Past"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\n\"Champion\" is a 1949 American drama film noir sports film directed by Mark Robson and starring an American actor whose other early films included a 1945 film noir directed by Jacques Tourneur.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Champion (1949 film)",
"Kirk Douglas",
"Out of the Past"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3424,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
23930
|
Titov, a Soviet cosmonaut born on September 11, 1935, was the first person to orbit the Earth multiple times (a total of 17), and was also the first to sleep in orbit and to suffer from space sickness.
|
{
"evidence": [
"Gherman Titov_sentence_1",
"Gherman Titov_sentence_5",
"Gherman Titov_sentence_6"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nTitov, a Soviet cosmonaut born on September 11, 1935, was the first person to orbit the Earth multiple times (a total of 17), and was also the first to sleep in orbit and to suffer from space sickness.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Gherman Titov_sentence_1",
"Gherman Titov_sentence_5",
"Gherman Titov_sentence_6"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3425,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.621",
"value": "?search="
},
{
"operation": "search",
"time": "12.887",
"value": "Titov"
},
{
"operation": "Now on",
"time": "13.536",
"value": "Titov"
},
{
"operation": "hyperlink",
"time": "21.059",
"value": "Gherman Titov"
},
{
"operation": "Now on",
"time": "21.893",
"value": "Gherman Titov"
},
{
"operation": "Highlighting",
"time": "26.118",
"value": "Gherman Titov_sentence_1"
},
{
"operation": "Highlighting",
"time": "39.381",
"value": "Gherman Titov_sentence_5"
},
{
"operation": "Highlighting",
"time": "44.874",
"value": "Gherman Titov_sentence_6"
},
{
"operation": "finish",
"time": "67.392",
"value": "finish"
}
]
}
|
41558
|
Coke R. Stevenson and Lyndon B. Johnson were regular civilians that only participated in school elections.
|
{
"evidence": [
"Coke R. Stevenson_sentence_0",
"Lyndon B. Johnson_sentence_4"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nCoke R. Stevenson and Lyndon B. Johnson were regular civilians that only participated in school elections.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Coke R. Stevenson_sentence_0",
"Lyndon B. Johnson_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3426,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.832",
"value": "?search="
},
{
"operation": "search",
"time": "13.629",
"value": "Coke R. Stevenson"
},
{
"operation": "Now on",
"time": "14.541",
"value": "Coke R. Stevenson"
},
{
"operation": "Highlighting",
"time": "43.683",
"value": "Coke R. Stevenson_sentence_0"
},
{
"operation": "search",
"time": "52.227",
"value": "Lyndon B. Johnson"
},
{
"operation": "Now on",
"time": "53.699",
"value": "Lyndon B. Johnson"
},
{
"operation": "Highlighting",
"time": "55.499",
"value": "Lyndon B. Johnson_sentence_4"
},
{
"operation": "finish",
"time": "66.161",
"value": "finish"
}
]
}
|
exfever_6208
|
Queens is the most populous borough in New York City.
|
{
"evidence": [
"New York City",
"New York metropolitan area"
],
"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:\nQueens is the most populous borough in New York City.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"New York City",
"New York metropolitan area"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3427,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3421
|
The Martian is a film genre that uses speculative, fictional science-based depictions of phenomena that are not fully accepted by mainstream science.
|
{
"evidence": [
"The Martian (film)",
"Science fiction 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:\nThe Martian is a film genre that uses speculative, fictional science-based depictions of phenomena that are not fully accepted by mainstream science.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Martian (film)",
"Science fiction film"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3428,
"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": 3429,
"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"
}
]
}
|
69001
|
The 1980 Austrian Grand Prix was a Formula One motor race which is the highest class of international auto racing for single-seater racing cars.
|
{
"evidence": [
"1980 Austrian Grand Prix_sentence_0",
"Formula One_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe 1980 Austrian Grand Prix was a Formula One motor race which is the highest class of international auto racing for single-seater racing cars.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"1980 Austrian Grand Prix_sentence_0",
"Formula One_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3430,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.19",
"value": "?search="
},
{
"operation": "search",
"time": "9.871",
"value": "1980 Austrian Grand Prix"
},
{
"operation": "Now on",
"time": "11.6",
"value": "1980 Austrian Grand Prix"
},
{
"operation": "Highlighting",
"time": "13.33",
"value": "1980 Austrian Grand Prix_sentence_0"
},
{
"operation": "hyperlink",
"time": "15.303",
"value": "Formula One"
},
{
"operation": "Now on",
"time": "18.172",
"value": "Formula One"
},
{
"operation": "Highlighting",
"time": "20.123",
"value": "Formula One_sentence_2"
},
{
"operation": "finish",
"time": "30.704",
"value": "finish"
}
]
}
|
exfever_6195
|
The Night Of is a 2016 American eight-part crime drama television miniseries written by an American screenwriter, director, film editor, and producer who won an Academy Award.
|
{
"evidence": [
"The Night Of",
"Steven Zaillian"
],
"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 Night Of is a 2016 American eight-part crime drama television miniseries written by an American screenwriter, director, film editor, and producer who won an Academy Award.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Night Of",
"Steven Zaillian"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3431,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
60802
|
The flag of Naples had a castle on it to honor its former kingdom.
|
{
"evidence": [
"Flag of Naples_sentence_1"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe flag of Naples had a castle on it to honor its former kingdom.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Flag of Naples_sentence_1"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3432,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.208",
"value": "?search="
},
{
"operation": "search",
"time": "6.134",
"value": "flag of Naples"
},
{
"operation": "Now on",
"time": "8.428",
"value": "Flag of Naples"
},
{
"operation": "Highlighting",
"time": "32.404",
"value": "Flag of Naples_sentence_1"
},
{
"operation": "finish",
"time": "94.03",
"value": "finish"
}
]
}
|
exfever_2326
|
Vantage Point is a political action thriller film that focuses on an assassination attempt on the President of the United States in Salamanca, in a country in southwestern Asia.
|
{
"evidence": [
"Vantage Point (film)",
"Spain"
],
"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:\nVantage Point is a political action thriller film that focuses on an assassination attempt on the President of the United States in Salamanca, in a country in southwestern Asia.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Vantage Point (film)",
"Spain"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3433,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
73169
|
In the Cartesian plane, two vectors are said to be perpendicular if the angle between them is 90°, while in Cartesian space, the norm of a vector is the square root of the vector dotted with itself.
|
{
"evidence": [
"Orthonormality_sentence_7",
"Orthonormality_sentence_4"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nIn the Cartesian plane, two vectors are said to be perpendicular if the angle between them is 90°, while in Cartesian space, the norm of a vector is the square root of the vector dotted with itself.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Orthonormality_sentence_7",
"Orthonormality_sentence_4"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3434,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.836",
"value": "?search="
},
{
"operation": "search",
"time": "12.729",
"value": "Cartesian"
},
{
"operation": "Now on",
"time": "13.573",
"value": "Cartesian"
},
{
"operation": "search",
"time": "30.594",
"value": "cartesian space"
},
{
"operation": "Now on",
"time": "32.246",
"value": "undefined"
},
{
"operation": "search",
"time": "80.115",
"value": "foot of a perpendicular"
},
{
"operation": "Now on",
"time": "81.431",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "84.124",
"value": "Perpendicular"
},
{
"operation": "Now on",
"time": "84.829",
"value": "Perpendicular"
},
{
"operation": "search",
"time": "121.301",
"value": "Orthonormality"
},
{
"operation": "Now on",
"time": "122.428",
"value": "Orthonormality"
},
{
"operation": "Highlighting",
"time": "125.067",
"value": "Orthonormality_sentence_7"
},
{
"operation": "Highlighting",
"time": "127.92",
"value": "Orthonormality_sentence_4"
},
{
"operation": "finish",
"time": "144.776",
"value": "finish"
}
]
}
|
66496
|
Active in showbusiness since 1958, Julia Roberts (born February 20, 1946) is an actress, comedian, singer, and dancer, known for her performances in the Broadway revival of Peter Pan and in the sitcom The Hogan Family, an American sitcom television series that began airing on NBC on March 1, 1986 and finished its run on CBS on July 20, 1991, for a total of six seasons.
|
{
"evidence": [
"Sandy Duncan_sentence_1",
"Sandy Duncan_sentence_2",
"The Hogan Family_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:\nActive in showbusiness since 1958, Julia Roberts (born February 20, 1946) is an actress, comedian, singer, and dancer, known for her performances in the Broadway revival of Peter Pan and in the sitcom The Hogan Family, an American sitcom television series that began airing on NBC on March 1, 1986 and finished its run on CBS on July 20, 1991, for a total of six seasons.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Sandy Duncan_sentence_1",
"Sandy Duncan_sentence_2",
"The Hogan Family_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3435,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.651",
"value": "?search="
},
{
"operation": "search",
"time": "11.948",
"value": "Julia Roberts"
},
{
"operation": "Now on",
"time": "14.555",
"value": "Julia Roberts"
},
{
"operation": "Highlighting",
"time": "28.313",
"value": "Julia Roberts_sentence_4"
},
{
"operation": "Highlighting",
"time": "28.314",
"value": "Julia Roberts_title"
},
{
"operation": "search",
"time": "47.669",
"value": "Broadway revival of Peter Pan and in the sitcom The Hogan Family"
},
{
"operation": "Now on",
"time": "52.17",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "54.437",
"value": "Television and video Sandy Duncan_section_4"
},
{
"operation": "Now on",
"time": "55.563",
"value": "Sandy Duncan#Television and video Sandy Duncan section 4"
},
{
"operation": "Highlighting",
"time": "59.709",
"value": "Sandy Duncan_sentence_1"
},
{
"operation": "Highlighting",
"time": "63.917",
"value": "Sandy Duncan_sentence_2"
},
{
"operation": "hyperlink",
"time": "65.99",
"value": "The Hogan Family"
},
{
"operation": "Now on",
"time": "67.702",
"value": "The Hogan Family"
},
{
"operation": "Highlighting",
"time": "69.033",
"value": "The Hogan Family_sentence_0"
},
{
"operation": "finish",
"time": "75.446",
"value": "finish"
}
]
}
|
3752
|
During Good Friday, some parishes of the Anglican, Catholic, and Lutheran Church the "Two Hours Devotion" is observed in which this traditionally consists sermons, and other mass-related elements.
|
{
"evidence": [
"Three Hours' Agony_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:\nDuring Good Friday, some parishes of the Anglican, Catholic, and Lutheran Church the \"Two Hours Devotion\" is observed in which this traditionally consists sermons, and other mass-related elements.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Three Hours' Agony_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3436,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "search",
"time": "28.241",
"value": "Good Friday"
},
{
"operation": "Page search",
"time": "39.482",
"value": "Two Hours Devotion"
},
{
"operation": "search",
"time": "57.816",
"value": "Two Hours Devotion"
},
{
"operation": "back-button-clicked",
"time": "83.903",
"value": "back-button-clicked"
},
{
"operation": "Page search",
"time": "89.456",
"value": "three hours devotion"
},
{
"operation": "hyperlink",
"time": "110.251",
"value": "Three Hours Devotion"
},
{
"operation": "Highlighting",
"time": "118.109",
"value": "Three Hours' Agony_sentence_0"
},
{
"operation": "back-button-clicked",
"time": "196.125",
"value": "back-button-clicked"
},
{
"operation": "hyperlink",
"time": "254.337",
"value": "Three Hours Devotion"
},
{
"operation": "search",
"time": "269.854",
"value": "two hours devotion"
},
{
"operation": "hyperlink",
"time": "283.54",
"value": "Ecce Homo (Martínez and Giménez)"
},
{
"operation": "Page search",
"time": "293.429",
"value": "two"
},
{
"operation": "finish",
"time": "308.677",
"value": "finish"
}
]
}
|
18495
|
Before Wrath of the Titans came out, reports came out that director Louis Leterrier would be the director while Greg Berlanti would write the story on the installment of the first film.
|
{
"evidence": [
"Wrath of the Titans_sentence_53"
],
"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:\nBefore Wrath of the Titans came out, reports came out that director Louis Leterrier would be the director while Greg Berlanti would write the story on the installment of the first film.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Wrath of the Titans_sentence_53"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3437,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.913",
"value": "undefined"
},
{
"operation": "search",
"time": "13.917",
"value": "Louis Leterrier"
},
{
"operation": "Now on",
"time": "14.889",
"value": "Louis Leterrier"
},
{
"operation": "search",
"time": "105.827",
"value": "Wrath of the Titans"
},
{
"operation": "Now on",
"time": "107.224",
"value": "Wrath of the Titans"
},
{
"operation": "Page search",
"time": "147.904",
"value": "louis"
},
{
"operation": "Highlighting",
"time": "164.489",
"value": "Wrath of the Titans_sentence_53"
},
{
"operation": "hyperlink",
"time": "241.028",
"value": "Clash of the Titans"
},
{
"operation": "Now on",
"time": "241.806",
"value": "Clash of the Titans (2010 film)"
},
{
"operation": "Page search",
"time": "276.721",
"value": "louis"
},
{
"operation": "back-button-clicked",
"time": "328.248",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "328.386",
"value": "Wrath of the Titans"
},
{
"operation": "finish",
"time": "345.148",
"value": "finish"
}
]
}
|
exfever_1402
|
Beautiful is a song recorded by an Asian singer and actress who signed with a record label owned by Sony Music Entertainment, a subsidiary of Sony Corporation of America.
|
{
"evidence": [
"Beautiful (Christina Aguilera song)",
"Christina Aguilera",
"RCA Records"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBeautiful is a song recorded by an Asian singer and actress who signed with a record label owned by Sony Music Entertainment, a subsidiary of Sony Corporation of America.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Beautiful (Christina Aguilera song)",
"Christina Aguilera",
"RCA Records"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3438,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5774
|
Legion was the directorial debut of Noah Hawley.
|
{
"evidence": [
"Legion (TV series)",
"Showrunner"
],
"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:\nLegion was the directorial debut of Noah Hawley.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Legion (TV series)",
"Showrunner"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3439,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3469
|
Kirk Douglas was an Indian actor and filmmaker, who became an international star for his role as an unscrupulous boxing hero in a 1949 American drama film noir sport film which pictures the struggles of "Midge" Kelly, an athlete in combat sport in which two people, usually wearing protective gloves and other protective equipment.
|
{
"evidence": [
"Kirk Douglas",
"Champion (1949 film)",
"Boxing"
],
"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:\nKirk Douglas was an Indian actor and filmmaker, who became an international star for his role as an unscrupulous boxing hero in a 1949 American drama film noir sport film which pictures the struggles of \"Midge\" Kelly, an athlete in combat sport in which two people, usually wearing protective gloves and other protective equipment.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Kirk Douglas",
"Champion (1949 film)",
"Boxing"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3440,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2306
|
Andhra Pradesh, India's seventh-largest state, is bordered by the state in India situated on the south-central stretch of the Indian peninsula on the high plateau that lies between the Western Ghats and the Eastern Ghats.
|
{
"evidence": [
"Andhra Pradesh",
"Telangana",
"Deccan Plateau"
],
"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:\nAndhra Pradesh, India's seventh-largest state, is bordered by the state in India situated on the south-central stretch of the Indian peninsula on the high plateau that lies between the Western Ghats and the Eastern Ghats.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Andhra Pradesh",
"Telangana",
"Deccan Plateau"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3441,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
69166
|
On the list of Vice Chairs for the North District Council, four of the Vice Chairmen have been affiliated with Heung Yee Kuk.
|
{
"evidence": [
"Heung Yee Kuk (constituency)_sentence_0",
"Heung Yee Kuk (constituency)_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:\nOn the list of Vice Chairs for the North District Council, four of the Vice Chairmen have been affiliated with Heung Yee Kuk.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Heung Yee Kuk (constituency)_sentence_0",
"Heung Yee Kuk (constituency)_sentence_1"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3442,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.229",
"value": "?search="
},
{
"operation": "search",
"time": "25.634",
"value": "North District Council Heung Yee Kuk"
},
{
"operation": "Now on",
"time": "28.185",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "216.113",
"value": "Cheung Yan-lung"
},
{
"operation": "Now on",
"time": "216.882",
"value": "Cheung Yan-lung"
},
{
"operation": "search",
"time": "242.481",
"value": "list of vice chairs for the North district council"
},
{
"operation": "Now on",
"time": "246.561",
"value": "undefined"
},
{
"operation": "search",
"time": "268.609",
"value": "Heung Yee Kuk"
},
{
"operation": "Now on",
"time": "270.804",
"value": "Heung Yee Kuk"
},
{
"operation": "hyperlink",
"time": "354.944",
"value": "Heung Yee Kuk (constituency)"
},
{
"operation": "Now on",
"time": "359.956",
"value": "Heung Yee Kuk (constituency)"
},
{
"operation": "Highlighting",
"time": "386.481",
"value": "Heung Yee Kuk (constituency)_sentence_0"
},
{
"operation": "Highlighting",
"time": "399.528",
"value": "Heung Yee Kuk (constituency)_sentence_1"
},
{
"operation": "search",
"time": "424.528",
"value": "North District Council china"
},
{
"operation": "Now on",
"time": "429.357",
"value": "undefined"
},
{
"operation": "finish",
"time": "445.32",
"value": "finish"
}
]
}
|
exfever_4970
|
Ronald Reagan previously served as the 33rd commander-in-chief of the California National Guard and the California State Guard.
|
{
"evidence": [
"Ronald Reagan",
"Governor of California"
],
"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:\nRonald Reagan previously served as the 33rd commander-in-chief of the California National Guard and the California State Guard.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Ronald Reagan",
"Governor of California"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3443,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
88995
|
The Japan Animator Expo is a weekly series of original net animations that was first announced by Hideaki Anno and was designed to offer new animators more exposure.
|
{
"evidence": [
"Japan Animator Expo_sentence_0",
"Japan Animator Expo_sentence_4",
"Japan Animator Expo_sentence_5"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Japan Animator Expo is a weekly series of original net animations that was first announced by Hideaki Anno and was designed to offer new animators more exposure.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Japan Animator Expo_sentence_0",
"Japan Animator Expo_sentence_4",
"Japan Animator Expo_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3444,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.611",
"value": "?search="
},
{
"operation": "search",
"time": "14.882",
"value": "Japan Animator Expo"
},
{
"operation": "Now on",
"time": "16.328",
"value": "Japan Animator Expo"
},
{
"operation": "Highlighting",
"time": "29.741",
"value": "Japan Animator Expo_sentence_0"
},
{
"operation": "Highlighting",
"time": "47.305",
"value": "Japan Animator Expo_sentence_4"
},
{
"operation": "Highlighting",
"time": "56.605",
"value": "Japan Animator Expo_sentence_5"
},
{
"operation": "finish",
"time": "75.374",
"value": "finish"
}
]
}
|
37913
|
The three Spanish athletes who competed at the 1954 European Athletics Championships in Bern, Switzerland, from 25–29 August 1954, did not advance to the semi-finals.
|
{
"evidence": [
"1954 European Athletics Championships_sentence_0"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe three Spanish athletes who competed at the 1954 European Athletics Championships in Bern, Switzerland, from 25–29 August 1954, did not advance to the semi-finals.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"1954 European Athletics Championships_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3445,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.657",
"value": "?search="
},
{
"operation": "search",
"time": "34.217",
"value": "1954 European Athletics Championships"
},
{
"operation": "Now on",
"time": "35.425",
"value": "1954 European Athletics Championships"
},
{
"operation": "Highlighting",
"time": "40.585",
"value": "1954 European Athletics Championships_sentence_0"
},
{
"operation": "search",
"time": "160.828",
"value": "1954 european athletics championship spain"
},
{
"operation": "Now on",
"time": "162.872",
"value": "undefined"
},
{
"operation": "search",
"time": "330.137",
"value": "he three Spanish athletes who competed at the 1954 European Athletics Championships in Bern, Switzerland, from 25–29 August 1954, did not advance to the semi-finals"
},
{
"operation": "Now on",
"time": "332.338",
"value": "undefined"
},
{
"operation": "search",
"time": "357.195",
"value": "spanish athletes 1954 european athletics"
},
{
"operation": "Now on",
"time": "358.018",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "371.924",
"value": "List of European Athletics Championships records"
},
{
"operation": "Now on",
"time": "373.043",
"value": "List of European Athletics Championships records"
},
{
"operation": "search",
"time": "398.235",
"value": "1954 European Athletics Championships – Men's 10,000 metres track walk"
},
{
"operation": "Now on",
"time": "399.138",
"value": "1954 European Athletics Championships – Men's 10,000 metres track walk"
},
{
"operation": "search",
"time": "434.308",
"value": "1954 european athletics championships semi finals"
},
{
"operation": "Now on",
"time": "435.437",
"value": "undefined"
},
{
"operation": "search",
"time": "454.939",
"value": "1954 European Athletics Championships"
},
{
"operation": "Now on",
"time": "455.678",
"value": "1954 European Athletics Championships"
},
{
"operation": "hyperlink",
"time": "474.217",
"value": "Soviet Union"
},
{
"operation": "Now on",
"time": "475.308",
"value": "Soviet Union"
},
{
"operation": "back-button-clicked",
"time": "477.236",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "478.712",
"value": "1954 European Athletics Championships"
},
{
"operation": "finish",
"time": "509.793",
"value": "finish"
}
]
}
|
exfever_7076
|
Radioactive is an electronic rock and alternative rock song with elements of electronic dance music that originated in South Korea in the early 2000s.
|
{
"evidence": [
"Radioactive (Imagine Dragons song)",
"Dubstep"
],
"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:\nRadioactive is an electronic rock and alternative rock song with elements of electronic dance music that originated in South Korea in the early 2000s.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Radioactive (Imagine Dragons song)",
"Dubstep"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3446,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_970
|
Billy Bob Thornton is an American actor, writer, and director, who starred in a 1998 American science fiction disaster film produced and directed by Michael Bay.
|
{
"evidence": [
"Billy Bob Thornton",
"Armageddon (1998 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:\nBilly Bob Thornton is an American actor, writer, and director, who starred in a 1998 American science fiction disaster film produced and directed by Michael Bay.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Billy Bob Thornton",
"Armageddon (1998 film)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3447,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
60164
|
In 2014 Fenbrook Institution merged with the Beaver Creek Institution, a minimum and medium security Canadian prison opened in 1961.
|
{
"evidence": [
"Fenbrook Institution_sentence_1",
"Beaver Creek Institution_sentence_0",
"Beaver Creek Institution_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:\nIn 2014 Fenbrook Institution merged with the Beaver Creek Institution, a minimum and medium security Canadian prison opened in 1961.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Fenbrook Institution_sentence_1",
"Beaver Creek Institution_sentence_0",
"Beaver Creek Institution_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3448,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.806",
"value": "?search="
},
{
"operation": "search",
"time": "12.298",
"value": "2014 Fenbrook Institution"
},
{
"operation": "Now on",
"time": "12.88",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "16.603",
"value": "Fenbrook Institution"
},
{
"operation": "Now on",
"time": "17.247",
"value": "Fenbrook Institution"
},
{
"operation": "Highlighting",
"time": "24.08",
"value": "Fenbrook Institution_sentence_1"
},
{
"operation": "hyperlink",
"time": "27.556",
"value": "Beaver Creek Institution"
},
{
"operation": "Now on",
"time": "28.034",
"value": "Beaver Creek Institution"
},
{
"operation": "Highlighting",
"time": "31.278",
"value": "Beaver Creek Institution_sentence_0"
},
{
"operation": "Highlighting",
"time": "280.29",
"value": "Beaver Creek Institution_sentence_2"
},
{
"operation": "finish",
"time": "297.064",
"value": "finish"
}
]
}
|
74932
|
The Western Area Peninsula National Park which has suffered from deforestation from urban encroachment, contains many endangered species.
|
{
"evidence": [
"Western Area Peninsula National Park_sentence_9",
"Western Area Peninsula National Park_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:\nThe Western Area Peninsula National Park which has suffered from deforestation from urban encroachment, contains many endangered species.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Western Area Peninsula National Park_sentence_9",
"Western Area Peninsula National Park_sentence_8"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3449,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.037",
"value": "?search="
},
{
"operation": "search",
"time": "14.072",
"value": "Western Area Peninsula National Park"
},
{
"operation": "Now on",
"time": "14.901",
"value": "Western Area Peninsula National Park"
},
{
"operation": "Page search",
"time": "23.622",
"value": "deforestation"
},
{
"operation": "Highlighting",
"time": "31.33",
"value": "Western Area Peninsula National Park_sentence_9"
},
{
"operation": "Page search",
"time": "40.995",
"value": "endangered"
},
{
"operation": "Highlighting",
"time": "46.7",
"value": "Western Area Peninsula National Park_sentence_8"
},
{
"operation": "finish",
"time": "55.947",
"value": "finish"
}
]
}
|
exfever_212
|
Weekly Idol is a South Korean variety show with its host for the first season being a South Korean comedian and television host for a South Korean television entertainment program.
|
{
"evidence": [
"Weekly Idol",
"Jeong Hyeong-don",
"Infinite Challenge"
],
"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:\nWeekly Idol is a South Korean variety show with its host for the first season being a South Korean comedian and television host for a South Korean television entertainment program.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Weekly Idol",
"Jeong Hyeong-don",
"Infinite Challenge"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3450,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7640
|
Ghostbusters grossed $229 million worldwide against its $144 million budget.
|
{
"evidence": [
"Ghostbusters (2016 film)",
"Comedy film"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nGhostbusters grossed $229 million worldwide against its $144 million budget.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Ghostbusters (2016 film)",
"Comedy film"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3451,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8302
|
Gia is an English biographical drama television film starring Faye Dunaway as a Dutch-American model who began with Ford Models.
|
{
"evidence": [
"Gia",
"Wilhelmina Cooper"
],
"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:\nGia is an English biographical drama television film starring Faye Dunaway as a Dutch-American model who began with Ford Models.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Gia",
"Wilhelmina Cooper"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3452,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8169
|
My Summer of Love is a 2004 British drama film filmed in a metropolitan and ceremonial county in Switzerland.
|
{
"evidence": [
"My Summer of Love",
"West Yorkshire"
],
"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:\nMy Summer of Love is a 2004 British drama film filmed in a metropolitan and ceremonial county in Switzerland.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"My Summer of Love",
"West Yorkshire"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3453,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1695
|
Enemy is a 2013 British psychological drama film that was loosely adapted from a 2002 novel by Portuguese author José Saramago, who won an award that is awarded to an author from any country who has produced the most outstanding work in an idealistic direction.
|
{
"evidence": [
"Enemy (2013 film)",
"The Double (Saramago novel)",
"Nobel Prize in Literature"
],
"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:\nEnemy is a 2013 British psychological drama film that was loosely adapted from a 2002 novel by Portuguese author José Saramago, who won an award that is awarded to an author from any country who has produced the most outstanding work in an idealistic direction.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Enemy (2013 film)",
"The Double (Saramago novel)",
"Nobel Prize in Literature"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3454,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
35727
|
Švenčionys Ghetto had 566 Jews - some of which established contracts with active Lithuanian policemen, who frustrated economical exploitation plans of the Germans towards Soviet territories, while the others joined Jewish partisans.
|
{
"evidence": [
"Švenčionys Ghetto_sentence_14",
"Švenčionys Ghetto_sentence_16",
"Soviet partisans_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:\nŠvenčionys Ghetto had 566 Jews - some of which established contracts with active Lithuanian policemen, who frustrated economical exploitation plans of the Germans towards Soviet territories, while the others joined Jewish partisans.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Švenčionys Ghetto_sentence_14",
"Švenčionys Ghetto_sentence_16",
"Soviet partisans_sentence_3"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3455,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.643",
"value": "?search="
},
{
"operation": "search",
"time": "13.53",
"value": "Svencionys Ghetto"
},
{
"operation": "Now on",
"time": "14.443",
"value": "Švenčionys Ghetto"
},
{
"operation": "Highlighting",
"time": "34.154",
"value": "Švenčionys Ghetto_sentence_14"
},
{
"operation": "Highlighting",
"time": "135.623",
"value": "Švenčionys Ghetto_sentence_16"
},
{
"operation": "hyperlink",
"time": "138.411",
"value": "Soviet partisans"
},
{
"operation": "Now on",
"time": "139.197",
"value": "Soviet partisans"
},
{
"operation": "Highlighting",
"time": "188.613",
"value": "Soviet partisans_sentence_3"
},
{
"operation": "finish",
"time": "262.06",
"value": "finish"
}
]
}
|
exfever_1167
|
Sparta was the biggest city in Laconia, in ancient Greece.
|
{
"evidence": [
"Sparta",
"Peloponnese"
],
"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:\nSparta was the biggest city in Laconia, in ancient Greece.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Sparta",
"Peloponnese"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3456,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
60839
|
Kohzoh Imai, known for the discovery of a melanoma-related antigen by producing a monoclonal antibody, had received her MD from a Japanese University in 1972 and Ph.D. five years later.
|
{
"evidence": [
"Kohzoh Imai_sentence_11",
"Kohzoh Imai_sentence_12"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nKohzoh Imai, known for the discovery of a melanoma-related antigen by producing a monoclonal antibody, had received her MD from a Japanese University in 1972 and Ph.D. five years later.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Kohzoh Imai_sentence_11",
"Kohzoh Imai_sentence_12"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3457,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.76",
"value": "?search="
},
{
"operation": "search",
"time": "13.174",
"value": "Kohzoh Imai"
},
{
"operation": "Now on",
"time": "14.335",
"value": "Kohzoh Imai"
},
{
"operation": "Highlighting",
"time": "22.718",
"value": "Kohzoh Imai_sentence_1"
},
{
"operation": "Page search",
"time": "45.684",
"value": "receive"
},
{
"operation": "Highlighting",
"time": "53.641",
"value": "Kohzoh Imai_sentence_11"
},
{
"operation": "Highlighting",
"time": "79.183",
"value": "Kohzoh Imai_sentence_12"
},
{
"operation": "search",
"time": "86.502",
"value": "Sapporo Medical College"
},
{
"operation": "Now on",
"time": "89.012",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "102.555",
"value": "History Chūō-ku, Sapporo_section_0"
},
{
"operation": "Now on",
"time": "103.306",
"value": "Chūō-ku, Sapporo#History Ch.C5.AB.C5.8D-ku.2C Sapporo section 0"
},
{
"operation": "Page search",
"time": "116.238",
"value": "Sapporo Medical College"
},
{
"operation": "Page search",
"time": "127.549",
"value": "medical"
},
{
"operation": "hyperlink",
"time": "140.594",
"value": "Sapporo Medical University"
},
{
"operation": "Now on",
"time": "141.25",
"value": "Sapporo Medical University"
},
{
"operation": "Highlighting",
"time": "152.197",
"value": "Sapporo Medical University_sentence_0"
},
{
"operation": "Highlighting",
"time": "152.198",
"value": "Sapporo Medical University_title"
},
{
"operation": "back-button-clicked",
"time": "154.069",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "154.242",
"value": "Chūō-ku, Sapporo#History Ch.C5.AB.C5.8D-ku.2C Sapporo section 0"
},
{
"operation": "back-button-clicked",
"time": "156.355",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "156.367",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "157.141",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "158.858",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "158.87",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "158.985",
"value": "Kohzoh Imai"
},
{
"operation": "finish",
"time": "403.415",
"value": "finish"
}
]
}
|
exfever_7524
|
Paris fatally wounds the son of Nereus and Doris' daughter.
|
{
"evidence": [
"Paris (mythology)",
"Achilles",
"Thetis"
],
"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:\nParis fatally wounds the son of Nereus and Doris' daughter.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Paris (mythology)",
"Achilles",
"Thetis"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3458,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5479
|
Whitney Houston is one of the best-selling recording artists of all time who sang a song originally recorded by Marilyn McCoo and Billy Davis Jr. as a cover song for her self-titled debut studio album.
|
{
"evidence": [
"Whitney Houston",
"Saving All My Love for You",
"Whitney Houston (album)"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nWhitney Houston is one of the best-selling recording artists of all time who sang a song originally recorded by Marilyn McCoo and Billy Davis Jr. as a cover song for her self-titled debut studio album.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Whitney Houston",
"Saving All My Love for You",
"Whitney Houston (album)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3459,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3581
|
South Park is an American animated series created by an American actor, animator, writer, producer, director, and composer born on October 19, 1969.
|
{
"evidence": [
"South Park",
"Trey Parker"
],
"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:\nSouth Park is an American animated series created by an American actor, animator, writer, producer, director, and composer born on October 19, 1969.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"South Park",
"Trey Parker"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3460,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
71165
|
The 2016 Six Nations Championship, known as the 2016 RBS 6 Nations due to the tournament's sponsorship by The Royal Bank of Scotland, was contested by England and France only.
|
{
"evidence": [
"2016 Six Nations Championship_sentence_1"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe 2016 Six Nations Championship, known as the 2016 RBS 6 Nations due to the tournament's sponsorship by The Royal Bank of Scotland, was contested by England and France only.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"2016 Six Nations Championship_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3461,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.915",
"value": "?search="
},
{
"operation": "search",
"time": "7.813",
"value": "2016 Six Nations Championship"
},
{
"operation": "Now on",
"time": "9.191",
"value": "2016 Six Nations Championship"
},
{
"operation": "Highlighting",
"time": "12.544",
"value": "2016 Six Nations Championship_sentence_0"
},
{
"operation": "Highlighting",
"time": "41.521",
"value": "2016 Six Nations Championship_sentence_1"
},
{
"operation": "finish",
"time": "53.324",
"value": "finish"
}
]
}
|
exfever_5412
|
Kirk Douglas made his debut film with an American actress, model and dancer.
|
{
"evidence": [
"Kirk Douglas",
"Barbara Stanwyck"
],
"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:\nKirk Douglas made his debut film with an American actress, model and dancer.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Kirk Douglas",
"Barbara Stanwyck"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3462,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
72415
|
Albert Einstein discovered the 1263 Varsavia in Uccle Obs.
|
{
"evidence": [
"1263 Varsavia_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:\nAlbert Einstein discovered the 1263 Varsavia in Uccle Obs.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"1263 Varsavia_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3463,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.047",
"value": "?search="
},
{
"operation": "search",
"time": "18.734",
"value": "Albert Einstein"
},
{
"operation": "Now on",
"time": "20.697",
"value": "Albert Einstein"
},
{
"operation": "search",
"time": "203.151",
"value": "1263 Varsavia"
},
{
"operation": "Now on",
"time": "205.299",
"value": "1263 Varsavia"
},
{
"operation": "Highlighting",
"time": "217.962",
"value": "1263 Varsavia_sentence_1"
},
{
"operation": "finish",
"time": "235.747",
"value": "finish"
}
]
}
|
80563
|
Standard Paper Box Corporation has a complex which consists of three buildings which are notable in the historical development of power systems for industrial facilities.
|
{
"evidence": [
"Standard Paper Box Corporation_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:\nStandard Paper Box Corporation has a complex which consists of three buildings which are notable in the historical development of power systems for industrial facilities.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Standard Paper Box Corporation_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3464,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.793",
"value": "?search="
},
{
"operation": "search",
"time": "58.057",
"value": "Standard Paper Box Corporation"
},
{
"operation": "Now on",
"time": "58.765",
"value": "Standard Paper Box Corporation"
},
{
"operation": "Highlighting",
"time": "60.87",
"value": "Standard Paper Box Corporation_sentence_0"
},
{
"operation": "Highlighting",
"time": "68.469",
"value": "Standard Paper Box Corporation_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "78.701",
"value": "Standard Paper Box Corporation_sentence_0"
},
{
"operation": "finish",
"time": "96.156",
"value": "finish"
}
]
}
|
exfever_7942
|
Michael Kelland John Hutchence recorded some solo material and acted in feature films, including a 1990 science fiction horror film based on Brian Aldiss' 1973 novel of the same name, starring an English actor whose career spanned over five decades.
|
{
"evidence": [
"Michael Hutchence",
"Frankenstein Unbound",
"John Hurt"
],
"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:\nMichael Kelland John Hutchence recorded some solo material and acted in feature films, including a 1990 science fiction horror film based on Brian Aldiss' 1973 novel of the same name, starring an English actor whose career spanned over five decades.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Michael Hutchence",
"Frankenstein Unbound",
"John Hurt"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3465,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
48056
|
The 20th Genie Awards were hosted by Patrick McKenna who is a Canadian comedian and actor who played Harold Green on The Red Green Show.
|
{
"evidence": [
"20th Genie Awards_sentence_1",
"Patrick McKenna_sentence_1",
"Patrick McKenna_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe 20th Genie Awards were hosted by Patrick McKenna who is a Canadian comedian and actor who played Harold Green on The Red Green Show.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"20th Genie Awards_sentence_1",
"Patrick McKenna_sentence_1",
"Patrick McKenna_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3466,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.98",
"value": "?search="
},
{
"operation": "search",
"time": "8.803",
"value": "20th Genie Awards"
},
{
"operation": "Now on",
"time": "9.677",
"value": "20th Genie Awards"
},
{
"operation": "Highlighting",
"time": "15.725",
"value": "20th Genie Awards_sentence_1"
},
{
"operation": "hyperlink",
"time": "18.028",
"value": "Patrick McKenna"
},
{
"operation": "Now on",
"time": "19.057",
"value": "Patrick McKenna"
},
{
"operation": "Highlighting",
"time": "24.381",
"value": "Patrick McKenna_sentence_1"
},
{
"operation": "Highlighting",
"time": "26.996",
"value": "Patrick McKenna_sentence_2"
},
{
"operation": "finish",
"time": "44.821",
"value": "finish"
}
]
}
|
58393
|
The Grand Ridge Rail Trail, a 12 km long rail trail in Mirboo North, Victoria, Australia, has easy trail difficulty and some travel on, and crossings of, major highways.
|
{
"evidence": [
"Grand Ridge Rail Trail_sentence_0"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Grand Ridge Rail Trail, a 12 km long rail trail in Mirboo North, Victoria, Australia, has easy trail difficulty and some travel on, and crossings of, major highways.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Grand Ridge Rail Trail_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3467,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.477",
"value": "?search="
},
{
"operation": "search",
"time": "10.167",
"value": "Grand Ridge Rail Trail"
},
{
"operation": "Now on",
"time": "11.305",
"value": "Grand Ridge Rail Trail"
},
{
"operation": "Highlighting",
"time": "39.201",
"value": "Grand Ridge Rail Trail_sentence_0"
},
{
"operation": "finish",
"time": "50.13",
"value": "finish"
}
]
}
|
exfever_2864
|
Tessanne Chin has opened for artists such as an American singer who is the winner of two awards presented by the Recording Academy to recognize "Outstanding Achievement in the Music Industry" in the United States.
|
{
"evidence": [
"Tessanne Chin",
"Peabo Bryson",
"Grammy Award"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nTessanne Chin has opened for artists such as an American singer who is the winner of two awards presented by the Recording Academy to recognize \"Outstanding Achievement in the Music Industry\" in the United States.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Tessanne Chin",
"Peabo Bryson",
"Grammy Award"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3468,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2461
|
Calvin Harris is a DJ and record producer who released his first studio album, 18 Months, which involves more dance music in a broad range of percussive electronic music genres made largely for nightclubs.
|
{
"evidence": [
"Calvin Harris",
"Electronic dance music"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nCalvin Harris is a DJ and record producer who released his first studio album, 18 Months, which involves more dance music in a broad range of percussive electronic music genres made largely for nightclubs.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Calvin Harris",
"Electronic dance music"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3469,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
85660
|
Babylon Mystery Religion did not sell well.
|
{
"evidence": [
"Babylon Mystery Religion_sentence_0",
"Babylon Mystery Religion_sentence_3",
"Babylon Mystery Religion_sentence_4"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBabylon Mystery Religion did not sell well.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Babylon Mystery Religion_sentence_0",
"Babylon Mystery Religion_sentence_3",
"Babylon Mystery Religion_sentence_4"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3470,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.817",
"value": "?search="
},
{
"operation": "search",
"time": "58.171",
"value": "Babylon Mystery Religion"
},
{
"operation": "Now on",
"time": "59.062",
"value": "Babylon Mystery Religion"
},
{
"operation": "search",
"time": "445.861",
"value": "Babylon Mystery Religion did not sell well"
},
{
"operation": "Now on",
"time": "448.885",
"value": "undefined"
},
{
"operation": "search",
"time": "463.61",
"value": "Contains...Babylon Mystery Religion"
},
{
"operation": "Now on",
"time": "464.762",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "481.813",
"value": "Cremation of Care"
},
{
"operation": "Now on",
"time": "482.562",
"value": "Cremation of Care"
},
{
"operation": "search",
"time": "544.943",
"value": "Contains...\"Babylon Mystery Religion\""
},
{
"operation": "Now on",
"time": "545.307",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "549.026",
"value": "Babylon Mystery Religion"
},
{
"operation": "Now on",
"time": "549.654",
"value": "Babylon Mystery Religion"
},
{
"operation": "Highlighting",
"time": "552.876",
"value": "Babylon Mystery Religion_sentence_0"
},
{
"operation": "Highlighting",
"time": "557.804",
"value": "Babylon Mystery Religion_sentence_3"
},
{
"operation": "Highlighting",
"time": "559.569",
"value": "Babylon Mystery Religion_sentence_4"
},
{
"operation": "finish",
"time": "577.167",
"value": "finish"
}
]
}
|
exfever_5880
|
Christopher Lloyd is the first American actor to receive a Drama Desk award.
|
{
"evidence": [
"Christopher Lloyd",
"Obie Award",
"American Theatre Wing"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nChristopher Lloyd is the first American actor to receive a Drama Desk award.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Christopher Lloyd",
"Obie Award",
"American Theatre Wing"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3471,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5600
|
Tezaab starred an Indian actor and film producer who received several accolades, including two of the most prominent film award ceremony in India.
|
{
"evidence": [
"Tezaab",
"Anil Kapoor",
"National Film Awards"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nTezaab starred an Indian actor and film producer who received several accolades, including two of the most prominent film award ceremony in India.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Tezaab",
"Anil Kapoor",
"National Film Awards"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3472,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
76881
|
Dr. Clara Immerwahr (21 June 1870 – 2 May 1915) was a German physicist.
|
{
"evidence": [
"Clara Immerwahr_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:\nDr. Clara Immerwahr (21 June 1870 – 2 May 1915) was a German physicist.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Clara Immerwahr_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3473,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.791",
"value": "?search="
},
{
"operation": "search",
"time": "13.168",
"value": "Contains...Dr. Clara Immerwahr"
},
{
"operation": "Now on",
"time": "13.976",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "16.107",
"value": "Clara Immerwahr"
},
{
"operation": "Now on",
"time": "16.639",
"value": "Clara Immerwahr"
},
{
"operation": "Highlighting",
"time": "19.275",
"value": "Clara Immerwahr_sentence_0"
},
{
"operation": "finish",
"time": "25.993",
"value": "finish"
}
]
}
|
29607
|
Anatoli Fyodorovich Polosin was never involved with football.
|
{
"evidence": [
"Anatoli Polosin_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:\nAnatoli Fyodorovich Polosin was never involved with football.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Anatoli Polosin_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3474,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.169",
"value": "?search="
},
{
"operation": "search",
"time": "27.768",
"value": "Contains...Anatoli Fyodorovich Polosin"
},
{
"operation": "Now on",
"time": "28.746",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "46.64",
"value": "Anatoli Polosin"
},
{
"operation": "Now on",
"time": "47.598",
"value": "Anatoli Polosin"
},
{
"operation": "Highlighting",
"time": "69.132",
"value": "Anatoli Polosin_sentence_0"
},
{
"operation": "finish",
"time": "77.331",
"value": "finish"
}
]
}
|
16685
|
Gabrovo Municipality, named after the Gabrovo Province (and its capital city, Gabrovo), is located in the Balkan Mountains, which house some of Europe's largest animals.
|
{
"evidence": [
"Gabrovo Municipality_sentence_0",
"Gabrovo Municipality_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:\nGabrovo Municipality, named after the Gabrovo Province (and its capital city, Gabrovo), is located in the Balkan Mountains, which house some of Europe's largest animals.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Gabrovo Municipality_sentence_0",
"Gabrovo Municipality_sentence_1"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3475,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.428",
"value": "?search="
},
{
"operation": "search",
"time": "21.607",
"value": "Gabrovo Municipality"
},
{
"operation": "Now on",
"time": "23.213",
"value": "Gabrovo Municipality"
},
{
"operation": "Highlighting",
"time": "27.588",
"value": "Gabrovo Municipality_header_cell_0_0_0"
},
{
"operation": "Highlighting deleted",
"time": "29.794",
"value": "Gabrovo Municipality_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "35.454",
"value": "Gabrovo Municipality_sentence_0"
},
{
"operation": "Highlighting",
"time": "43.903",
"value": "Gabrovo Municipality_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "58.44",
"value": "Gabrovo Municipality_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "87.551",
"value": "Gabrovo Municipality_sentence_1"
},
{
"operation": "Highlighting",
"time": "88.204",
"value": "Gabrovo Municipality_sentence_0"
},
{
"operation": "Highlighting",
"time": "88.816",
"value": "Gabrovo Municipality_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "159.966",
"value": "Gabrovo Municipality_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "160.817",
"value": "Gabrovo Municipality_sentence_1"
},
{
"operation": "Highlighting",
"time": "179.212",
"value": "Gabrovo Municipality_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "196.695",
"value": "Gabrovo Municipality_sentence_1"
},
{
"operation": "Highlighting",
"time": "198.722",
"value": "Gabrovo Municipality_sentence_0"
},
{
"operation": "hyperlink",
"time": "200.589",
"value": "Stara planina"
},
{
"operation": "Now on",
"time": "202.937",
"value": "Balkan Mountains"
},
{
"operation": "back-button-clicked",
"time": "204.603",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "204.863",
"value": "Gabrovo Municipality"
},
{
"operation": "Highlighting",
"time": "206.25",
"value": "Gabrovo Municipality_sentence_1"
},
{
"operation": "hyperlink",
"time": "221.4",
"value": "Gabrovo"
},
{
"operation": "Now on",
"time": "222.944",
"value": "Gabrovo"
},
{
"operation": "finish",
"time": "238.545",
"value": "finish"
}
]
}
|
exfever_5402
|
All My Children is the last television soap opera aired on ABC.
|
{
"evidence": [
"All My Children",
"Rosemont, Pennsylvania"
],
"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:\nAll My Children is the last television soap opera aired on ABC.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"All My Children",
"Rosemont, Pennsylvania"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3476,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
22292
|
There was not even one Mauthausen Orchestra track created between 1986 and 1987.
|
{
"evidence": [
"Mauthausen Orchestra_sentence_6"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThere was not even one Mauthausen Orchestra track created between 1986 and 1987.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Mauthausen Orchestra_sentence_6"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3477,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.671",
"value": "?search="
},
{
"operation": "search",
"time": "13.157",
"value": "Mauthausen Orchestra"
},
{
"operation": "Now on",
"time": "14.752",
"value": "Mauthausen Orchestra"
},
{
"operation": "Highlighting",
"time": "25.259",
"value": "Mauthausen Orchestra_sentence_6"
},
{
"operation": "finish",
"time": "29.714",
"value": "finish"
}
]
}
|
19119
|
The Nerodia floridana is a venomous water snake (also known as the Florida Green Watersnake), classified in the Animalia Kingdom, under the class known as Reptilia.
|
{
"evidence": [
"Nerodia floridana_sentence_0",
"Nerodia_sentence_0"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Nerodia floridana is a venomous water snake (also known as the Florida Green Watersnake), classified in the Animalia Kingdom, under the class known as Reptilia.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Nerodia floridana_sentence_0",
"Nerodia_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3478,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.543",
"value": "?search="
},
{
"operation": "search",
"time": "37.171",
"value": "Nerodia floridana"
},
{
"operation": "Now on",
"time": "38.324",
"value": "Nerodia floridana"
},
{
"operation": "Highlighting",
"time": "83.795",
"value": "Nerodia floridana_sentence_0"
},
{
"operation": "Page search",
"time": "99.623",
"value": "venom"
},
{
"operation": "hyperlink",
"time": "113.985",
"value": "water snakes"
},
{
"operation": "Now on",
"time": "114.944",
"value": "Nerodia"
},
{
"operation": "Highlighting",
"time": "117.021",
"value": "Nerodia_sentence_0"
},
{
"operation": "back-button-clicked",
"time": "118.115",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "118.318",
"value": "Nerodia floridana"
},
{
"operation": "finish",
"time": "129.111",
"value": "finish"
}
]
}
|
32143
|
Isan people overran the Lao Wang and forced them into corvee Iintermitten unpaid labour expected by landowners or monarchs) labour before they were integrated into Isan farming; much of Isan was populated this way.
|
{
"evidence": [
"Isan people_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:\nIsan people overran the Lao Wang and forced them into corvee Iintermitten unpaid labour expected by landowners or monarchs) labour before they were integrated into Isan farming; much of Isan was populated this way.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Isan people_sentence_1"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3479,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.688",
"value": "?search="
},
{
"operation": "search",
"time": "19.68",
"value": "Isan people"
},
{
"operation": "Now on",
"time": "20.685",
"value": "Isan people"
},
{
"operation": "search",
"time": "55.599",
"value": "lao wang"
},
{
"operation": "Now on",
"time": "57.403",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "63.42",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "63.547",
"value": "Isan people"
},
{
"operation": "Page search",
"time": "69.941",
"value": "lao wang"
},
{
"operation": "search",
"time": "90.143",
"value": "Lao wang"
},
{
"operation": "Now on",
"time": "90.763",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "102.588",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "102.718",
"value": "Isan people"
},
{
"operation": "search",
"time": "114.804",
"value": "overran"
},
{
"operation": "Now on",
"time": "115.534",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "124.556",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "124.699",
"value": "Isan people"
},
{
"operation": "Page search",
"time": "129.652",
"value": "overran"
},
{
"operation": "Highlighting",
"time": "133.105",
"value": "Isan people_sentence_1"
},
{
"operation": "search",
"time": "167.422",
"value": "san people overran the Lao Wang"
},
{
"operation": "Now on",
"time": "168.509",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "175.966",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "176.084",
"value": "Isan people"
},
{
"operation": "search",
"time": "197.322",
"value": "corvee Iintermitten unpaid labour"
},
{
"operation": "Now on",
"time": "198.335",
"value": "undefined"
},
{
"operation": "search",
"time": "211.559",
"value": "Isan people"
},
{
"operation": "Now on",
"time": "212.187",
"value": "Isan people"
},
{
"operation": "finish",
"time": "228.97",
"value": "finish"
}
]
}
|
75289
|
Chaco Slope, noted for its Chacoan Anasazi ruins and first labelled by archaeologist Gwinn Vivian, borders the Chaco Core; which contains both Chacra Mesa and Chaco Canyon.
|
{
"evidence": [
"Chaco Slope_sentence_3",
"Chaco Slope_sentence_5",
"Chaco Slope_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:\nChaco Slope, noted for its Chacoan Anasazi ruins and first labelled by archaeologist Gwinn Vivian, borders the Chaco Core; which contains both Chacra Mesa and Chaco Canyon.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Chaco Slope_sentence_3",
"Chaco Slope_sentence_5",
"Chaco Slope_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3480,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.357",
"value": "?search="
},
{
"operation": "Now on",
"time": "2.589",
"value": "?search="
},
{
"operation": "search",
"time": "4.666",
"value": "Chaco Slope"
},
{
"operation": "Now on",
"time": "5.196",
"value": "Chaco Slope"
},
{
"operation": "Highlighting",
"time": "9.656",
"value": "Chaco Slope_sentence_3"
},
{
"operation": "Highlighting",
"time": "17.853",
"value": "Chaco Slope_sentence_5"
},
{
"operation": "Highlighting",
"time": "22.344",
"value": "Chaco Slope_sentence_2"
},
{
"operation": "finish",
"time": "33.599",
"value": "finish"
}
]
}
|
exfever_8663
|
Rosario Dawson won an Academy Award for the film Unforgettable.
|
{
"evidence": [
"Unforgettable (2017 film)",
"Cheryl Ladd"
],
"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:\nRosario Dawson won an Academy Award for the film Unforgettable.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Unforgettable (2017 film)",
"Cheryl Ladd"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3481,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
82410
|
Tyler Sash, born on May 27, 1988 in Oskaloosa, Iowa, played as safety for Oskaloosa High School and Iowa College, and was drafted neither by the Giants nor by the Falcons in the sixth round of the 2011 NFL Draft.
|
{
"evidence": [
"Tyler Sash_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:\nTyler Sash, born on May 27, 1988 in Oskaloosa, Iowa, played as safety for Oskaloosa High School and Iowa College, and was drafted neither by the Giants nor by the Falcons in the sixth round of the 2011 NFL Draft.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Tyler Sash_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3482,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.797",
"value": "?search="
},
{
"operation": "search",
"time": "12.972",
"value": "Tyler Sash"
},
{
"operation": "Now on",
"time": "13.984",
"value": "Tyler Sash"
},
{
"operation": "Highlighting",
"time": "17.422",
"value": "Tyler Sash_header_cell_0_3_0"
},
{
"operation": "Highlighting",
"time": "18.272",
"value": "Tyler Sash_cell_0_3_1"
},
{
"operation": "Highlighting",
"time": "18.282",
"value": "Tyler Sash_header_cell_0_3_0"
},
{
"operation": "Highlighting",
"time": "18.282",
"value": "Tyler Sash_header_cell_0_2_0"
},
{
"operation": "Highlighting",
"time": "30.864",
"value": "Tyler Sash_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "31.424",
"value": "Tyler Sash_cell_0_1_1"
},
{
"operation": "Highlighting",
"time": "31.435",
"value": "Tyler Sash_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "31.435",
"value": "Tyler Sash_header_cell_0_0_0"
},
{
"operation": "Highlighting deleted",
"time": "42.732",
"value": "Tyler Sash_cell_0_1_1"
},
{
"operation": "Highlighting deleted",
"time": "42.741",
"value": "Tyler Sash_header_cell_0_1_0"
},
{
"operation": "Highlighting deleted",
"time": "42.743",
"value": "Tyler Sash_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "58.195",
"value": "Tyler Sash_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "58.748",
"value": "Tyler Sash_cell_0_1_1"
},
{
"operation": "Highlighting",
"time": "58.759",
"value": "Tyler Sash_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "58.76",
"value": "Tyler Sash_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "60.88",
"value": "Tyler Sash_header_cell_0_8_0"
},
{
"operation": "Highlighting",
"time": "62.345",
"value": "Tyler Sash_cell_0_8_1"
},
{
"operation": "Highlighting",
"time": "62.357",
"value": "Tyler Sash_header_cell_0_8_0"
},
{
"operation": "Highlighting",
"time": "62.357",
"value": "Tyler Sash_header_cell_0_7_0"
},
{
"operation": "Highlighting",
"time": "64.67",
"value": "Tyler Sash_header_cell_0_9_0"
},
{
"operation": "Highlighting",
"time": "65.287",
"value": "Tyler Sash_cell_0_9_1"
},
{
"operation": "Highlighting",
"time": "65.3",
"value": "Tyler Sash_header_cell_0_9_0"
},
{
"operation": "Highlighting deleted",
"time": "93.116",
"value": "Tyler Sash_cell_0_3_1"
},
{
"operation": "Highlighting deleted",
"time": "93.125",
"value": "Tyler Sash_header_cell_0_3_0"
},
{
"operation": "Highlighting deleted",
"time": "93.126",
"value": "Tyler Sash_header_cell_0_2_0"
},
{
"operation": "Highlighting deleted",
"time": "94.148",
"value": "Tyler Sash_cell_0_1_1"
},
{
"operation": "Highlighting deleted",
"time": "94.161",
"value": "Tyler Sash_header_cell_0_1_0"
},
{
"operation": "Highlighting deleted",
"time": "94.162",
"value": "Tyler Sash_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "95.417",
"value": "Tyler Sash_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "97.892",
"value": "Tyler Sash_cell_0_8_1"
},
{
"operation": "Highlighting deleted",
"time": "97.901",
"value": "Tyler Sash_header_cell_0_8_0"
},
{
"operation": "Highlighting deleted",
"time": "98.379",
"value": "Tyler Sash_cell_0_9_1"
},
{
"operation": "Highlighting deleted",
"time": "98.389",
"value": "Tyler Sash_header_cell_0_9_0"
},
{
"operation": "Highlighting deleted",
"time": "98.391",
"value": "Tyler Sash_header_cell_0_7_0"
},
{
"operation": "finish",
"time": "127.583",
"value": "finish"
}
]
}
|
18338
|
Mönchengladbach Mavericks' fortunes changed from 2007, they were promoted in 2008 and entered the German Football League 2 which is the second tier of American football in Germany.
|
{
"evidence": [
"Mönchengladbach Mavericks_sentence_6",
"Mönchengladbach Mavericks_sentence_7",
"Mönchengladbach Mavericks_sentence_8",
"German Football League 2_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:\nMönchengladbach Mavericks' fortunes changed from 2007, they were promoted in 2008 and entered the German Football League 2 which is the second tier of American football in Germany.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Mönchengladbach Mavericks_sentence_6",
"Mönchengladbach Mavericks_sentence_7",
"Mönchengladbach Mavericks_sentence_8",
"German Football League 2_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3483,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.088",
"value": "?search="
},
{
"operation": "search",
"time": "7.699",
"value": "Mönchengladbach Mavericks"
},
{
"operation": "Now on",
"time": "8.643",
"value": "Mönchengladbach Mavericks"
},
{
"operation": "Highlighting",
"time": "14.809",
"value": "Mönchengladbach Mavericks_sentence_6"
},
{
"operation": "Highlighting",
"time": "20.003",
"value": "Mönchengladbach Mavericks_sentence_7"
},
{
"operation": "Highlighting",
"time": "32.641",
"value": "Mönchengladbach Mavericks_sentence_8"
},
{
"operation": "hyperlink",
"time": "46.435",
"value": "German Football League 2"
},
{
"operation": "Now on",
"time": "47.507",
"value": "German Football League 2"
},
{
"operation": "Highlighting",
"time": "49.138",
"value": "German Football League 2_sentence_0"
},
{
"operation": "back-button-clicked",
"time": "50.811",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "50.888",
"value": "Mönchengladbach Mavericks"
},
{
"operation": "finish",
"time": "63.566",
"value": "finish"
}
]
}
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.