id
stringlengths 1
12
| question
stringlengths 22
566
| golden_answers
dict | data_source
stringclasses 2
values | prompt
listlengths 1
1
| ability
stringclasses 6
values | reward_model
dict | extra_info
dict | metadata
dict |
|---|---|---|---|---|---|---|---|---|
exfever_3908
|
My Everything is the album by Ariana Grande, released by a New York City-based Asian record label owned by Universal Music Group.
|
{
"evidence": [
"My Everything (Ariana Grande album)",
"Republic 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:\nMy Everything is the album by Ariana Grande, released by a New York City-based Asian record label owned by Universal Music Group.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"My Everything (Ariana Grande album)",
"Republic Records"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6684,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
55098
|
After overcoming his alcoholic addiction, Eric Clapton realized that he had wasted five years of his life, and later on he liked the recordings.
|
{
"evidence": [
"Eric Clapton_sentence_18",
"Eric Clapton_sentence_177",
"Eric Clapton_sentence_170",
"Eric Clapton_sentence_171"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nAfter overcoming his alcoholic addiction, Eric Clapton realized that he had wasted five years of his life, and later on he liked the recordings.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Eric Clapton_sentence_18",
"Eric Clapton_sentence_177",
"Eric Clapton_sentence_170",
"Eric Clapton_sentence_171"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6685,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.252",
"value": "undefined"
},
{
"operation": "search",
"time": "22.403",
"value": "Eric Clapton"
},
{
"operation": "Now on",
"time": "23.264",
"value": "Eric Clapton"
},
{
"operation": "Page search",
"time": "74.73",
"value": "wasted 5 years of his life"
},
{
"operation": "Page search",
"time": "106.137",
"value": "wasted"
},
{
"operation": "Page search",
"time": "137.736",
"value": "recordings"
},
{
"operation": "search",
"time": "359.678",
"value": "later on he liked the recordings"
},
{
"operation": "Now on",
"time": "362.464",
"value": "undefined"
},
{
"operation": "search",
"time": "383.838",
"value": "realized he had wasted 5 years of his life"
},
{
"operation": "Now on",
"time": "387.062",
"value": "undefined"
},
{
"operation": "search",
"time": "412.012",
"value": "eric clapton"
},
{
"operation": "Now on",
"time": "413.504",
"value": "Eric Clapton"
},
{
"operation": "search",
"time": "425.868",
"value": "overcoming"
},
{
"operation": "Now on",
"time": "426.485",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "428.932",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "429.102",
"value": "Eric Clapton"
},
{
"operation": "Page search",
"time": "432.422",
"value": "overcoming"
},
{
"operation": "Highlighting",
"time": "446.324",
"value": "Eric Clapton_sentence_18"
},
{
"operation": "hyperlink",
"time": "462.291",
"value": "Eric Clapton albums discography"
},
{
"operation": "Now on",
"time": "464.711",
"value": "Eric Clapton albums discography"
},
{
"operation": "back-button-clicked",
"time": "487.362",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "487.529",
"value": "Eric Clapton"
},
{
"operation": "hyperlink",
"time": "488.809",
"value": "Eric Clapton singles discography"
},
{
"operation": "Now on",
"time": "492.827",
"value": "Eric Clapton singles discography"
},
{
"operation": "back-button-clicked",
"time": "500.993",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "501.173",
"value": "Eric Clapton"
},
{
"operation": "Page search",
"time": "515.371",
"value": "alchohol"
},
{
"operation": "Page search",
"time": "520.106",
"value": "alcohol"
},
{
"operation": "Highlighting",
"time": "539.816",
"value": "Eric Clapton_sentence_177"
},
{
"operation": "Highlighting",
"time": "564.231",
"value": "Eric Clapton_sentence_170"
},
{
"operation": "Highlighting",
"time": "567.119",
"value": "Eric Clapton_sentence_171"
},
{
"operation": "finish",
"time": "614.853",
"value": "finish"
}
]
}
|
exfever_7586
|
Toy Story 3 is a 2010 American computer-animated comedy-drama film written by an American screenwriter best known for his works on a 2006 American tragicomedy road film and the feature film directorial debut of the husband–wife team of Jonathan Dayton and Valerie Faris.
|
{
"evidence": [
"Toy Story 3",
"Michael Arndt",
"Little Miss Sunshine"
],
"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:\nToy Story 3 is a 2010 American computer-animated comedy-drama film written by an American screenwriter best known for his works on a 2006 American tragicomedy road film and the feature film directorial debut of the husband–wife team of Jonathan Dayton and Valerie Faris.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Toy Story 3",
"Michael Arndt",
"Little Miss Sunshine"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6686,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7280
|
Lymelife is a 2009 American comedy-drama film produced by an American film director, producer, and screenwriter who is known for directing the film Who's That Knocking at My Door.
|
{
"evidence": [
"Lymelife",
"Martin Scorsese"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nLymelife is a 2009 American comedy-drama film produced by an American film director, producer, and screenwriter who is known for directing the film Who's That Knocking at My Door.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Lymelife",
"Martin Scorsese"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6687,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7693
|
Zeus is the Greek thunder god, which is the Roman equivalent of Jove, the god of the sky.
|
{
"evidence": [
"Zeus",
"Jupiter (mythology)"
],
"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:\nZeus is the Greek thunder god, which is the Roman equivalent of Jove, the god of the sky.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Zeus",
"Jupiter (mythology)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6688,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2449
|
Neil Patrick Harris has received multiple setbacks throughout his career including The Antoinette Perry Award for Excellence in Broadway Theatre.
|
{
"evidence": [
"Neil Patrick Harris",
"Tony Award"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nNeil Patrick Harris has received multiple setbacks throughout his career including The Antoinette Perry Award for Excellence in Broadway Theatre.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Neil Patrick Harris",
"Tony Award"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6689,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8383
|
The Reader is a 2008 romantic film with a category of narrative fiction intended to be more serious than humorous in tone.
|
{
"evidence": [
"The Reader (2008 film)",
"Drama (film and television)"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Reader is a 2008 romantic film with a category of narrative fiction intended to be more serious than humorous in tone.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Reader (2008 film)",
"Drama (film and television)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6690,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
23142
|
Bitee, nephew of Petro Atlético head coach Lazare Adingono, played on the Cameroonian men's national basketball team that won the silver medal at the FIBA Africa Championship 2007.
|
{
"evidence": [
"Parfait Bitee_sentence_3",
"Parfait Bitee_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:\nBitee, nephew of Petro Atlético head coach Lazare Adingono, played on the Cameroonian men's national basketball team that won the silver medal at the FIBA Africa Championship 2007.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Parfait Bitee_sentence_3",
"Parfait Bitee_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6691,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.837",
"value": "undefined"
},
{
"operation": "search",
"time": "21.322",
"value": "Contains...Cameroonian men's national basketball team"
},
{
"operation": "Now on",
"time": "23.646",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "28.12",
"value": "Parfait Bitee"
},
{
"operation": "Now on",
"time": "29.021",
"value": "Parfait Bitee"
},
{
"operation": "Highlighting",
"time": "35.207",
"value": "Parfait Bitee_sentence_3"
},
{
"operation": "Highlighting",
"time": "45.47",
"value": "Parfait Bitee_sentence_2"
},
{
"operation": "finish",
"time": "85.11",
"value": "finish"
}
]
}
|
29597
|
The 1931 Michigan State Normal Hurons football team played six games during the season and has lost three out of the six games.
|
{
"evidence": [
"1931 Michigan State Normal Hurons football team_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 1931 Michigan State Normal Hurons football team played six games during the season and has lost three out of the six games.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"1931 Michigan State Normal Hurons football team_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6692,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.705",
"value": "?search="
},
{
"operation": "search",
"time": "10.374",
"value": "1931 Michigan State Normal Hurons football"
},
{
"operation": "Now on",
"time": "11.088",
"value": "1931 Michigan State Normal Hurons football team"
},
{
"operation": "Highlighting",
"time": "41.039",
"value": "1931 Michigan State Normal Hurons football team_sentence_1"
},
{
"operation": "finish",
"time": "56.445",
"value": "finish"
}
]
}
|
24250
|
The left Front, like the Communist Party of India(Marxist), among others, won their most number of seats in Indian general elections in the 14th Lok Sabha elections.
|
{
"evidence": [
"Left Front (Tripura)_sentence_0",
"Left Front (Tripura)_sentence_2",
"Left Front (Tripura)_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:\nThe left Front, like the Communist Party of India(Marxist), among others, won their most number of seats in Indian general elections in the 14th Lok Sabha elections.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Left Front (Tripura)_sentence_0",
"Left Front (Tripura)_sentence_2",
"Left Front (Tripura)_sentence_3"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6693,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.058",
"value": "?search="
},
{
"operation": "search",
"time": "64.707",
"value": "left Front"
},
{
"operation": "Now on",
"time": "66.005",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "69.455",
"value": "Left Front"
},
{
"operation": "Now on",
"time": "70.371",
"value": "Left Front"
},
{
"operation": "hyperlink",
"time": "98.592",
"value": "Left Front (Tripura)"
},
{
"operation": "Now on",
"time": "99.546",
"value": "Left Front (Tripura)"
},
{
"operation": "Page search",
"time": "125.011",
"value": "Lok Sabha"
},
{
"operation": "Highlighting",
"time": "159.064",
"value": "Left Front (Tripura)_sentence_0"
},
{
"operation": "Highlighting",
"time": "162.919",
"value": "Left Front (Tripura)_sentence_2"
},
{
"operation": "Highlighting",
"time": "164.118",
"value": "Left Front (Tripura)_sentence_3"
},
{
"operation": "finish",
"time": "176.989",
"value": "finish"
}
]
}
|
11695
|
The two separate qualifying tournaments in the new system is the Champions Path, for clubs which won their domestic league and did not automatically qualify for the group stage, and the Non-Champions Path, for clubs which did not win their domestic league and did not automatically qualify for the group stage.
|
{
"evidence": [
"UEFA Women's Champions League_sentence_34"
],
"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 two separate qualifying tournaments in the new system is the Champions Path, for clubs which won their domestic league and did not automatically qualify for the group stage, and the Non-Champions Path, for clubs which did not win their domestic league and did not automatically qualify for the group stage.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"UEFA Women's Champions League_sentence_34"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6694,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.943",
"value": "?search="
},
{
"operation": "search",
"time": "26.802",
"value": "Champions Path"
},
{
"operation": "Now on",
"time": "34.826",
"value": "undefined"
},
{
"operation": "search",
"time": "77.715",
"value": "non-champions path"
},
{
"operation": "Now on",
"time": "80.512",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "86.026",
"value": "2010–11 UEFA Champions League group stage"
},
{
"operation": "Now on",
"time": "91.164",
"value": "2010–11 UEFA Champions League group stage"
},
{
"operation": "Highlighting",
"time": "111.55",
"value": "2010–11 UEFA Champions League group stage_sentence_0"
},
{
"operation": "Highlighting",
"time": "111.551",
"value": "2010–11 UEFA Champions League group stage_title"
},
{
"operation": "search",
"time": "118.314",
"value": "UEFA Champions League"
},
{
"operation": "Now on",
"time": "129.98",
"value": "UEFA Champions League"
},
{
"operation": "Page search",
"time": "155.03",
"value": "champions path"
},
{
"operation": "Page search",
"time": "182.198",
"value": "non-champions path"
},
{
"operation": "search",
"time": "287.626",
"value": "non-champions path"
},
{
"operation": "Now on",
"time": "292.797",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "305.218",
"value": "UEFA Women's Champions League"
},
{
"operation": "Now on",
"time": "311.099",
"value": "UEFA Women's Champions League"
},
{
"operation": "Page search",
"time": "328.98",
"value": "champions path"
},
{
"operation": "Page search",
"time": "347.411",
"value": "non-champions path"
},
{
"operation": "Highlighting",
"time": "380.067",
"value": "UEFA Women's Champions League_sentence_34"
},
{
"operation": "Now on",
"time": "468.181",
"value": "UEFA Women's Champions League"
},
{
"operation": "finish",
"time": "484.266",
"value": "finish"
}
]
}
|
45994
|
Miriam Meckel served as the State Secretary, she was a professor of communication sciences and wrote textbooks on cyberpolitics.
|
{
"evidence": [
"Miriam Meckel_sentence_3",
"Miriam Meckel_sentence_4",
"Miriam Meckel_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:\nMiriam Meckel served as the State Secretary, she was a professor of communication sciences and wrote textbooks on cyberpolitics.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Miriam Meckel_sentence_3",
"Miriam Meckel_sentence_4",
"Miriam Meckel_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6695,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.982",
"value": "?search="
},
{
"operation": "search",
"time": "6.134",
"value": "Miriam Mecel"
},
{
"operation": "Now on",
"time": "7.187",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "9.289",
"value": "miriam meckel"
},
{
"operation": "Now on",
"time": "9.944",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "15.92",
"value": "Miriam Meckel"
},
{
"operation": "Now on",
"time": "16.823",
"value": "Miriam Meckel"
},
{
"operation": "Highlighting",
"time": "23.202",
"value": "Miriam Meckel_sentence_3"
},
{
"operation": "Highlighting",
"time": "33.977",
"value": "Miriam Meckel_sentence_4"
},
{
"operation": "Highlighting",
"time": "37.823",
"value": "Miriam Meckel_sentence_5"
},
{
"operation": "finish",
"time": "51.463",
"value": "finish"
}
]
}
|
69914
|
PS 17 is a two-story, five-bay brick building on a high basement that features a shallow wooden entrance porch with Doric order(one of the three orders of ancient Greek and later Roman architecture) columns.
|
{
"evidence": [
"PS 17_sentence_4",
"PS 17_sentence_5",
"Doric order_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:\nPS 17 is a two-story, five-bay brick building on a high basement that features a shallow wooden entrance porch with Doric order(one of the three orders of ancient Greek and later Roman architecture) columns.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"PS 17_sentence_4",
"PS 17_sentence_5",
"Doric order_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6696,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.888",
"value": "?search="
},
{
"operation": "search",
"time": "15.85",
"value": "PS 17"
},
{
"operation": "Now on",
"time": "17.676",
"value": "PS 17"
},
{
"operation": "Highlighting",
"time": "21.072",
"value": "PS 17_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "23.215",
"value": "PS 17_sentence_0"
},
{
"operation": "Highlighting",
"time": "26.463",
"value": "PS 17_sentence_4"
},
{
"operation": "Highlighting",
"time": "33.015",
"value": "PS 17_sentence_5"
},
{
"operation": "hyperlink",
"time": "40.055",
"value": "Doric order"
},
{
"operation": "Now on",
"time": "40.75",
"value": "Doric order"
},
{
"operation": "Highlighting",
"time": "46.489",
"value": "Doric order_sentence_0"
},
{
"operation": "finish",
"time": "184.926",
"value": "finish"
}
]
}
|
exfever_4409
|
Geraldine Leigh Chaplin is an American actress who came to prominence in her portrayal of Tonya in a 1965 epic historical romance film directed by Will Smith.
|
{
"evidence": [
"Geraldine Chaplin",
"Doctor Zhivago (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:\nGeraldine Leigh Chaplin is an American actress who came to prominence in her portrayal of Tonya in a 1965 epic historical romance film directed by Will Smith.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Geraldine Chaplin",
"Doctor Zhivago (film)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6697,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4024
|
The Little Prince is a 2015 film directed by a Turkish film director, writer, producer, and animator.
|
{
"evidence": [
"The Little Prince (2015 film)",
"Mark Osborne (filmmaker)"
],
"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 Little Prince is a 2015 film directed by a Turkish film director, writer, producer, and animator.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Little Prince (2015 film)",
"Mark Osborne (filmmaker)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6698,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
48629
|
Highly Evolved is an album by The Vines that included the debut single "Highly Evolved", and placed 4th in the best album of the year race by NME (a British music, film and culture website and brand which has been around since 1952), as well as being included in the book 1001 Albums You Must Hear Before You Die.
|
{
"evidence": [
"Highly Evolved_sentence_0",
"Highly Evolved_sentence_5",
"NME_sentence_2"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nHighly Evolved is an album by The Vines that included the debut single \"Highly Evolved\", and placed 4th in the best album of the year race by NME (a British music, film and culture website and brand which has been around since 1952), as well as being included in the book 1001 Albums You Must Hear Before You Die.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Highly Evolved_sentence_0",
"Highly Evolved_sentence_5",
"NME_sentence_2"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6699,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.892",
"value": "?search="
},
{
"operation": "search",
"time": "10.186",
"value": "Highly Evolved"
},
{
"operation": "Now on",
"time": "11.007",
"value": "Highly Evolved"
},
{
"operation": "Highlighting",
"time": "15.156",
"value": "Highly Evolved_sentence_0"
},
{
"operation": "Highlighting",
"time": "25.117",
"value": "Highly Evolved_sentence_5"
},
{
"operation": "hyperlink",
"time": "89.063",
"value": "NME"
},
{
"operation": "Now on",
"time": "89.653",
"value": "NME"
},
{
"operation": "Highlighting",
"time": "101.159",
"value": "NME_sentence_2"
},
{
"operation": "Page search",
"time": "122.016",
"value": "year"
},
{
"operation": "hyperlink",
"time": "182.13",
"value": "NME Awards"
},
{
"operation": "Now on",
"time": "182.805",
"value": "NME Awards"
},
{
"operation": "Page search",
"time": "207.097",
"value": "vines"
},
{
"operation": "finish",
"time": "224.163",
"value": "finish"
}
]
}
|
50351
|
Freeland High school campus, housing grades ninth to twelfth, includes a football stadium, a soccer stadium, and a 500 person auditorium.
|
{
"evidence": [
"Freeland Community School District_sentence_9",
"Freeland Community School District_sentence_10"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nFreeland High school campus, housing grades ninth to twelfth, includes a football stadium, a soccer stadium, and a 500 person auditorium.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Freeland Community School District_sentence_9",
"Freeland Community School District_sentence_10"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6700,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "5.154",
"value": "?search="
},
{
"operation": "search",
"time": "435.551",
"value": "Freeland High School"
},
{
"operation": "Now on",
"time": "439.138",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "590.292",
"value": "Tommy Matchick"
},
{
"operation": "Now on",
"time": "591.308",
"value": "Tommy Matchick"
},
{
"operation": "search",
"time": "790.252",
"value": "Freeland Community School District"
},
{
"operation": "Now on",
"time": "791.1",
"value": "Freeland Community School District"
},
{
"operation": "Highlighting",
"time": "891.372",
"value": "Freeland Community School District_sentence_9"
},
{
"operation": "Highlighting",
"time": "917.355",
"value": "Freeland Community School District_sentence_10"
},
{
"operation": "finish",
"time": "1027.41",
"value": "finish"
}
]
}
|
exfever_1340
|
The Caspian Sea was the inspiration for the name of Prince Caspian in Narnia.
|
{
"evidence": [
"Caspian Sea",
"Eastern Europe"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Caspian Sea was the inspiration for the name of Prince Caspian in Narnia.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Caspian Sea",
"Eastern Europe"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6701,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8234
|
Annie is a contemporary transfer of a work to a short film.
|
{
"evidence": [
"Annie (2014 film)",
"Film adaptation"
],
"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:\nAnnie is a contemporary transfer of a work to a short film.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Annie (2014 film)",
"Film adaptation"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6702,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1101
|
Atomic Blonde is a 2017 Scottish action thriller film based on the 2012 graphic novel The Coldest City, starring a Scottish actor who made his acting debut as a teen in The Near Room.
|
{
"evidence": [
"Atomic Blonde",
"James McAvoy"
],
"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:\nAtomic Blonde is a 2017 Scottish action thriller film based on the 2012 graphic novel The Coldest City, starring a Scottish actor who made his acting debut as a teen in The Near Room.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Atomic Blonde",
"James McAvoy"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6703,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
84769
|
Conrad Schools of Science made a "Retire the Mascot Committee" against the Redskins mascot which had been used for 90 years.
|
{
"evidence": [
"Conrad Schools of Science_sentence_7",
"Conrad Schools of Science_sentence_8"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nConrad Schools of Science made a \"Retire the Mascot Committee\" against the Redskins mascot which had been used for 90 years.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Conrad Schools of Science_sentence_7",
"Conrad Schools of Science_sentence_8"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6704,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.085",
"value": "?search="
},
{
"operation": "search",
"time": "9.848",
"value": "Conrad Schools of Science"
},
{
"operation": "Now on",
"time": "10.524",
"value": "Conrad Schools of Science"
},
{
"operation": "Highlighting",
"time": "26.722",
"value": "Conrad Schools of Science_sentence_7"
},
{
"operation": "Highlighting",
"time": "28.445",
"value": "Conrad Schools of Science_sentence_8"
},
{
"operation": "finish",
"time": "44.98",
"value": "finish"
}
]
}
|
exfever_5309
|
Berlin is Germany's municipality holding secondary status in a country, state, province, department, or other sub-national entity, usually as its seat of the government.
|
{
"evidence": [
"Berlin",
"Capital city"
],
"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:\nBerlin is Germany's municipality holding secondary status in a country, state, province, department, or other sub-national entity, usually as its seat of the government.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Berlin",
"Capital city"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6705,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4480
|
Linda McCartney was an American photographer and musician who, along with Paul, after the 1970 breakup of the Beatles, recorded the only album credited to the husband-and-wife music duo, released on June 18, 1971, by Apple Records.
|
{
"evidence": [
"Linda McCartney",
"Ram (album)"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nLinda McCartney was an American photographer and musician who, along with Paul, after the 1970 breakup of the Beatles, recorded the only album credited to the husband-and-wife music duo, released on June 18, 1971, by Apple Records.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Linda McCartney",
"Ram (album)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6706,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
14772
|
PCFCL represents about 55% to 60% of primary cutaneous B-cell lymphomas (PCBCL) with unknown cause.
|
{
"evidence": [
"Primary cutaneous follicle center lymphoma_sentence_5",
"Primary cutaneous follicle center lymphoma_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:\nPCFCL represents about 55% to 60% of primary cutaneous B-cell lymphomas (PCBCL) with unknown cause.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Primary cutaneous follicle center lymphoma_sentence_5",
"Primary cutaneous follicle center lymphoma_sentence_6"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6707,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.48",
"value": "undefined"
},
{
"operation": "search",
"time": "32.71",
"value": "pcbcl"
},
{
"operation": "Now on",
"time": "33.216",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "36.566",
"value": "Primary cutaneous follicle center lymphoma"
},
{
"operation": "Now on",
"time": "37.142",
"value": "Primary cutaneous follicle center lymphoma"
},
{
"operation": "Highlighting",
"time": "56.07",
"value": "Primary cutaneous follicle center lymphoma_sentence_5"
},
{
"operation": "Highlighting",
"time": "57.069",
"value": "Primary cutaneous follicle center lymphoma_sentence_6"
},
{
"operation": "finish",
"time": "71.652",
"value": "finish"
}
]
}
|
exfever_5379
|
Shonen Jump was a shōnen manga anthology published in North America by an American manga publisher, anime distributor, and entertainment company headquartered in San Diego, California that was founded in 1986.
|
{
"evidence": [
"Shonen Jump (magazine)",
"Viz Media"
],
"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:\nShonen Jump was a shōnen manga anthology published in North America by an American manga publisher, anime distributor, and entertainment company headquartered in San Diego, California that was founded in 1986.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Shonen Jump (magazine)",
"Viz Media"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6708,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
60349
|
2002 Humboldt State Lumberjacks football team led by head coach Max Adkins and played home games at the Redwood Bowl in Arcata, California
|
{
"evidence": [
"2002 Humboldt State Lumberjacks football team_sentence_2"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\n2002 Humboldt State Lumberjacks football team led by head coach Max Adkins and played home games at the Redwood Bowl in Arcata, California\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"2002 Humboldt State Lumberjacks football team_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6709,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.748",
"value": "?search="
},
{
"operation": "search",
"time": "19.57",
"value": "2002 Humboldt State Lumberjacks football team"
},
{
"operation": "Now on",
"time": "20.107",
"value": "2002 Humboldt State Lumberjacks football team"
},
{
"operation": "Highlighting",
"time": "27.818",
"value": "2002 Humboldt State Lumberjacks football team_sentence_2"
},
{
"operation": "search",
"time": "48.474",
"value": "max adkins"
},
{
"operation": "Now on",
"time": "49.491",
"value": "undefined"
},
{
"operation": "finish",
"time": "165.348",
"value": "finish"
}
]
}
|
23278
|
Caroline Pidcock ( born in Grafton, New South Wales, on the Clarence River) is a campaigner for environmental and community issues, which led to her standing for the Legislative Council in the 2007 New South Wales state election.
|
{
"evidence": [
"Caroline Pidcock_sentence_1",
"Caroline Pidcock_sentence_2",
"Caroline Pidcock_sentence_9"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nCaroline Pidcock ( born in Grafton, New South Wales, on the Clarence River) is a campaigner for environmental and community issues, which led to her standing for the Legislative Council in the 2007 New South Wales state election.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Caroline Pidcock_sentence_1",
"Caroline Pidcock_sentence_2",
"Caroline Pidcock_sentence_9"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6710,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.69",
"value": "?search="
},
{
"operation": "search",
"time": "7.326",
"value": "Caroline Pidcock"
},
{
"operation": "Now on",
"time": "8.033",
"value": "Caroline Pidcock"
},
{
"operation": "Highlighting",
"time": "22.339",
"value": "Caroline Pidcock_sentence_1"
},
{
"operation": "Highlighting",
"time": "33.32",
"value": "Caroline Pidcock_sentence_2"
},
{
"operation": "Highlighting",
"time": "39.43",
"value": "Caroline Pidcock_sentence_9"
},
{
"operation": "finish",
"time": "44.974",
"value": "finish"
}
]
}
|
exfever_2394
|
Bohemian Rhapsody is a song by the British rock band Queen that has a loosely defined subgenre within the broad genre of popular music that originated as "rock and roll" in the United States.
|
{
"evidence": [
"Bohemian Rhapsody",
"Hard rock",
"Rock music"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBohemian Rhapsody is a song by the British rock band Queen that has a loosely defined subgenre within the broad genre of popular music that originated as \"rock and roll\" in the United States.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Bohemian Rhapsody",
"Hard rock",
"Rock music"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6711,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
41467
|
California AM radio station KROP started as a music station, took a talk show format and finally became a country music station.
|
{
"evidence": [
"KROP_sentence_7",
"KROP_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:\nCalifornia AM radio station KROP started as a music station, took a talk show format and finally became a country music station.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"KROP_sentence_7",
"KROP_sentence_6"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6712,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.213",
"value": "?search="
},
{
"operation": "Now on",
"time": "79.03",
"value": "?search="
},
{
"operation": "search",
"time": "91.25",
"value": "KROP"
},
{
"operation": "Now on",
"time": "92.048",
"value": "KROP"
},
{
"operation": "Highlighting",
"time": "127.711",
"value": "KROP_sentence_7"
},
{
"operation": "Highlighting",
"time": "128.464",
"value": "KROP_sentence_6"
},
{
"operation": "finish",
"time": "176.033",
"value": "finish"
}
]
}
|
exfever_7969
|
Oscar Palmer Robertson is a former professional basketball player who became an 11-time member of the annual National Basketball Association best players in the league following every NBA season.
|
{
"evidence": [
"Oscar Robertson",
"All-NBA Team"
],
"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:\nOscar Palmer Robertson is a former professional basketball player who became an 11-time member of the annual National Basketball Association best players in the league following every NBA season.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Oscar Robertson",
"All-NBA Team"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6713,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
19900
|
Knut Haakonsson's father served as steward of the Norwegian kingdom while his maternal grandfather reigned as King of Sweden in the 12th century.
|
{
"evidence": [
"Knut Haakonsson_sentence_2",
"Knut Haakonsson_sentence_3",
"Eric IX of Sweden_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:\nKnut Haakonsson's father served as steward of the Norwegian kingdom while his maternal grandfather reigned as King of Sweden in the 12th century.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Knut Haakonsson_sentence_2",
"Knut Haakonsson_sentence_3",
"Eric IX of Sweden_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6714,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.477",
"value": "?search="
},
{
"operation": "search",
"time": "22.925",
"value": "Knut Kaakonsson"
},
{
"operation": "Now on",
"time": "23.388",
"value": "undefined"
},
{
"operation": "search",
"time": "28.18",
"value": "knut"
},
{
"operation": "Now on",
"time": "28.741",
"value": "Knut"
},
{
"operation": "search",
"time": "55.428",
"value": "Haakonsson"
},
{
"operation": "Now on",
"time": "55.947",
"value": "Haakonsson"
},
{
"operation": "hyperlink",
"time": "60.1",
"value": "Knut Haakonsson"
},
{
"operation": "Now on",
"time": "60.496",
"value": "Knut Haakonsson"
},
{
"operation": "Highlighting",
"time": "75.189",
"value": "Knut Haakonsson_sentence_2"
},
{
"operation": "Highlighting",
"time": "92.54",
"value": "Knut Haakonsson_sentence_3"
},
{
"operation": "hyperlink",
"time": "93.348",
"value": "Eric IX of Sweden"
},
{
"operation": "Now on",
"time": "93.987",
"value": "Eric IX of Sweden"
},
{
"operation": "Highlighting",
"time": "97.236",
"value": "Eric IX of Sweden_sentence_0"
},
{
"operation": "finish",
"time": "103.732",
"value": "finish"
}
]
}
|
exfever_464
|
George Orson Welles was an Indian director who co-founded an independent repertory theatre company that presented a series of productions on Broadway through 1941, including a film that premiered on November 11, 1937, as the first production of the independent repertory theatre company founded in New York City in 1937 by Orson Welles.
|
{
"evidence": [
"Orson Welles",
"Caesar (Mercury Theatre)",
"Mercury Theatre"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nGeorge Orson Welles was an Indian director who co-founded an independent repertory theatre company that presented a series of productions on Broadway through 1941, including a film that premiered on November 11, 1937, as the first production of the independent repertory theatre company founded in New York City in 1937 by Orson Welles.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Orson Welles",
"Caesar (Mercury Theatre)",
"Mercury Theatre"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6715,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
14246
|
Israel was born to George, an engineer at Robins Air Force Base, and Margaret Israel, a teacher who helped veterans returning from World War II, in rural Bibb County, Georgia.
|
{
"evidence": [
"George Israel_sentence_4",
"George Israel_sentence_3"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nIsrael was born to George, an engineer at Robins Air Force Base, and Margaret Israel, a teacher who helped veterans returning from World War II, in rural Bibb County, Georgia.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"George Israel_sentence_4",
"George Israel_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6716,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.834",
"value": "undefined"
},
{
"operation": "search",
"time": "20.335",
"value": "Contains...Margaret Israel"
},
{
"operation": "Now on",
"time": "21.683",
"value": "undefined"
},
{
"operation": "search",
"time": "36.049",
"value": "\"Margaret Israel\""
},
{
"operation": "Now on",
"time": "36.544",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "38.646",
"value": "George Israel"
},
{
"operation": "Now on",
"time": "39.612",
"value": "George Israel"
},
{
"operation": "Highlighting",
"time": "48.054",
"value": "George Israel_sentence_3"
},
{
"operation": "Highlighting",
"time": "52.26",
"value": "George Israel_sentence_4"
},
{
"operation": "Highlighting",
"time": "69.894",
"value": "George Israel_cell_0_5_1"
},
{
"operation": "Highlighting",
"time": "69.903",
"value": "George Israel_header_cell_0_5_0"
},
{
"operation": "Highlighting",
"time": "69.903",
"value": "George Israel_header_cell_0_4_0"
},
{
"operation": "hyperlink",
"time": "71.309",
"value": "Bibb County, Georgia"
},
{
"operation": "Now on",
"time": "72.319",
"value": "Bibb County, Georgia"
},
{
"operation": "Page search",
"time": "82.82",
"value": "rural"
},
{
"operation": "Highlighting",
"time": "105.619",
"value": "Bibb County, Georgia_sentence_3"
},
{
"operation": "Highlighting",
"time": "105.62",
"value": "Bibb County, Georgia_title"
},
{
"operation": "back-button-clicked",
"time": "114.629",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "114.742",
"value": "George Israel"
},
{
"operation": "Page search",
"time": "118.374",
"value": "rural"
},
{
"operation": "Page search",
"time": "119.254",
"value": "rural"
},
{
"operation": "Highlighting deleted",
"time": "122.866",
"value": "George Israel_sentence_3"
},
{
"operation": "Highlighting",
"time": "123.6",
"value": "George Israel_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "137.66",
"value": "George Israel_cell_0_5_1"
},
{
"operation": "Highlighting deleted",
"time": "137.667",
"value": "George Israel_header_cell_0_5_0"
},
{
"operation": "Highlighting deleted",
"time": "137.669",
"value": "George Israel_header_cell_0_4_0"
},
{
"operation": "finish",
"time": "155.751",
"value": "finish"
}
]
}
|
exfever_3191
|
The Masked Raider was the first fictional character published by Marvel Comics.
|
{
"evidence": [
"Jean Grey",
"Mutant (Marvel Comics)",
"Puberty"
],
"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 Masked Raider was the first fictional character published by Marvel Comics.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Jean Grey",
"Mutant (Marvel Comics)",
"Puberty"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6717,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6737
|
Views is a studio album released by an African record label imprint of Cash Money Records and distributed by Republic Records.
|
{
"evidence": [
"Views (album)",
"Young Money Entertainment"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nViews is a studio album released by an African record label imprint of Cash Money Records and distributed by Republic Records.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Views (album)",
"Young Money Entertainment"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6718,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
85279
|
Blues hooker Royce Simmons typified the newfound New South Wales' spirit when early in the opening game he was knocked senseless; he staggered back into position, falling over and regaining his feet then waving trainers away, as he took his place in the defensive formation, before pulling off a try-saving tackle on his own tryline soon after.
|
{
"evidence": [
"Royce Simmons_sentence_6",
"Royce Simmons_sentence_8"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBlues hooker Royce Simmons typified the newfound New South Wales' spirit when early in the opening game he was knocked senseless; he staggered back into position, falling over and regaining his feet then waving trainers away, as he took his place in the defensive formation, before pulling off a try-saving tackle on his own tryline soon after.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Royce Simmons_sentence_6",
"Royce Simmons_sentence_8"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6719,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.155",
"value": "?search="
},
{
"operation": "search",
"time": "156.878",
"value": "Royce Simmons"
},
{
"operation": "Now on",
"time": "157.799",
"value": "Royce Simmons"
},
{
"operation": "Page search",
"time": "216.325",
"value": "blues"
},
{
"operation": "Highlighting",
"time": "228.679",
"value": "Royce Simmons_sentence_6"
},
{
"operation": "Page search",
"time": "255.277",
"value": "knocked senseless"
},
{
"operation": "Page search",
"time": "271.476",
"value": "knocked"
},
{
"operation": "Page search",
"time": "317.724",
"value": "stagger"
},
{
"operation": "Page search",
"time": "326.132",
"value": "staggered"
},
{
"operation": "search",
"time": "453.721",
"value": "Contains...Royce Simmons"
},
{
"operation": "Now on",
"time": "455.268",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "473.28",
"value": "1991 NSWRL season"
},
{
"operation": "Now on",
"time": "474.21",
"value": "1991 NSWRL season"
},
{
"operation": "Page search",
"time": "495.894",
"value": "staggered back"
},
{
"operation": "back-button-clicked",
"time": "559.884",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "560.443",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "582.822",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "582.825",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "582.87",
"value": "Royce Simmons"
},
{
"operation": "Highlighting",
"time": "592.306",
"value": "Royce Simmons_sentence_8"
},
{
"operation": "finish",
"time": "637.105",
"value": "finish"
}
]
}
|
72278
|
Sophia joined her father in Baltimore, contributing to the Southern Review and beginning a school for girls and she became a frequent contributor of articles, writing as Mrs S. B. Herrick, and was for a time assistant editor to Richard Watson Gilder at Century.
|
{
"evidence": [
"Sophia Bledsoe Herrick_sentence_3",
"Sophia Bledsoe Herrick_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:\nSophia joined her father in Baltimore, contributing to the Southern Review and beginning a school for girls and she became a frequent contributor of articles, writing as Mrs S. B. Herrick, and was for a time assistant editor to Richard Watson Gilder at Century.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Sophia Bledsoe Herrick_sentence_3",
"Sophia Bledsoe Herrick_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6720,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.077",
"value": "?search="
},
{
"operation": "search",
"time": "20.575",
"value": "Contains...Mrs S. B. Herrick"
},
{
"operation": "Now on",
"time": "22.943",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "26.11",
"value": "Sophia Bledsoe Herrick"
},
{
"operation": "Now on",
"time": "26.761",
"value": "Sophia Bledsoe Herrick"
},
{
"operation": "Highlighting",
"time": "33.27",
"value": "Sophia Bledsoe Herrick_sentence_3"
},
{
"operation": "Highlighting",
"time": "42.091",
"value": "Sophia Bledsoe Herrick_sentence_5"
},
{
"operation": "Page search",
"time": "54.39",
"value": "Gilder"
},
{
"operation": "finish",
"time": "77.5",
"value": "finish"
}
]
}
|
24540
|
Kungshatt is a small island in Sweden, and its name after a hat put in place on a cliff facing towards Vårberg.
|
{
"evidence": [
"Kungshatt_sentence_0",
"Kungshatt_sentence_3"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nKungshatt is a small island in Sweden, and its name after a hat put in place on a cliff facing towards Vårberg.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Kungshatt_sentence_0",
"Kungshatt_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6721,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.057",
"value": "?search="
},
{
"operation": "search",
"time": "193.633",
"value": "Kungshatt"
},
{
"operation": "Now on",
"time": "194.455",
"value": "Kungshatt"
},
{
"operation": "Highlighting",
"time": "264.341",
"value": "Kungshatt_sentence_0"
},
{
"operation": "Highlighting",
"time": "352.639",
"value": "Kungshatt_sentence_3"
},
{
"operation": "finish",
"time": "364.255",
"value": "finish"
}
]
}
|
62588
|
The Saint Louis Blue and White Football team took a break during the month of July in 1904.
|
{
"evidence": [
"1904 Saint Louis Blue and White football team_sentence_0"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Saint Louis Blue and White Football team took a break during the month of July in 1904.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"1904 Saint Louis Blue and White football team_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6722,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.432",
"value": "?search="
},
{
"operation": "search",
"time": "14.219",
"value": "Saint Louis Blue and White Football team"
},
{
"operation": "Now on",
"time": "16.935",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "31.108",
"value": "1904 Saint Louis Blue and White football team"
},
{
"operation": "Now on",
"time": "31.816",
"value": "1904 Saint Louis Blue and White football team"
},
{
"operation": "hyperlink",
"time": "58.337",
"value": "1904 college football season"
},
{
"operation": "Now on",
"time": "59.638",
"value": "1904 college football season"
},
{
"operation": "search",
"time": "75.631",
"value": "Saint Louis Blue and White Football team"
},
{
"operation": "Now on",
"time": "76.826",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "79.919",
"value": "1904 Saint Louis Blue and White football team"
},
{
"operation": "Now on",
"time": "81.014",
"value": "1904 Saint Louis Blue and White football team"
},
{
"operation": "Highlighting",
"time": "85.58",
"value": "1904 Saint Louis Blue and White football team_sentence_0"
},
{
"operation": "finish",
"time": "96.263",
"value": "finish"
}
]
}
|
63064
|
The House of Representatives is part of the legislative branch of the federal government of the United States consisting of congressional district representatives serving six-year terms.
|
{
"evidence": [
"United States Congress_sentence_8"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe House of Representatives is part of the legislative branch of the federal government of the United States consisting of congressional district representatives serving six-year terms.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"United States Congress_sentence_8"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6723,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.087",
"value": "?search="
},
{
"operation": "search",
"time": "16.211",
"value": "House of Representatives"
},
{
"operation": "Now on",
"time": "18",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "21.881",
"value": "United States House of Representatives"
},
{
"operation": "Now on",
"time": "23.359",
"value": "United States House of Representatives"
},
{
"operation": "Page search",
"time": "36.136",
"value": "legislative"
},
{
"operation": "Page search",
"time": "62.57",
"value": "branch"
},
{
"operation": "hyperlink",
"time": "100.372",
"value": "United States Congress"
},
{
"operation": "Now on",
"time": "101.904",
"value": "United States Congress"
},
{
"operation": "Page search",
"time": "109.753",
"value": "branch"
},
{
"operation": "hyperlink",
"time": "122.225",
"value": "separation of powers"
},
{
"operation": "Now on",
"time": "124.892",
"value": "Separation of powers"
},
{
"operation": "Page search",
"time": "129.582",
"value": "house"
},
{
"operation": "hyperlink",
"time": "169.792",
"value": "Separation of powers (disambiguation)"
},
{
"operation": "Now on",
"time": "170.428",
"value": "Separation of powers (disambiguation)"
},
{
"operation": "back-button-clicked",
"time": "174.628",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "174.876",
"value": "Separation of powers"
},
{
"operation": "back-button-clicked",
"time": "186.311",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "186.319",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "186.62",
"value": "United States Congress"
},
{
"operation": "back-button-clicked",
"time": "190.056",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "190.066",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "190.313",
"value": "United States House of Representatives"
},
{
"operation": "search",
"time": "202.389",
"value": "history of the house of representatives"
},
{
"operation": "Now on",
"time": "204.782",
"value": "undefined"
},
{
"operation": "search",
"time": "222.107",
"value": "Contains...House of Representatives is part of the legislative branch"
},
{
"operation": "Now on",
"time": "224.921",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "231.758",
"value": "United States Congress"
},
{
"operation": "Now on",
"time": "232.235",
"value": "United States Congress"
},
{
"operation": "Page search",
"time": "238.962",
"value": "legislative"
},
{
"operation": "Page search",
"time": "261.209",
"value": "legislative branch"
},
{
"operation": "back-button-clicked",
"time": "277.512",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "278.336",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "294.367",
"value": "Federal government of the United States"
},
{
"operation": "Now on",
"time": "295.642",
"value": "Federal government of the United States"
},
{
"operation": "Page search",
"time": "302.245",
"value": "legislative branch"
},
{
"operation": "Highlighting",
"time": "326.112",
"value": "Federal government of the United States_sentence_25"
},
{
"operation": "Highlighting",
"time": "326.113",
"value": "Federal government of the United States_title"
},
{
"operation": "Highlighting",
"time": "326.114",
"value": "Federal government of the United States_section_2"
},
{
"operation": "hyperlink",
"time": "328.089",
"value": "United States Congress"
},
{
"operation": "Now on",
"time": "328.825",
"value": "United States Congress"
},
{
"operation": "Page search",
"time": "335.941",
"value": "house of"
},
{
"operation": "Highlighting",
"time": "343.616",
"value": "United States Congress_sentence_2"
},
{
"operation": "Page search",
"time": "359.817",
"value": "district"
},
{
"operation": "Highlighting",
"time": "403.864",
"value": "United States Congress_sentence_8"
},
{
"operation": "finish",
"time": "464.659",
"value": "finish"
}
]
}
|
20017
|
Diana Dell was born in Grozny the capital city of Chechnya, Russia, The city lies on the Sunzha River.
|
{
"evidence": [
"Diana Dell_sentence_1",
"Grozny_sentence_1",
"Grozny_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:\nDiana Dell was born in Grozny the capital city of Chechnya, Russia, The city lies on the Sunzha River.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Diana Dell_sentence_1",
"Grozny_sentence_1",
"Grozny_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6724,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.992",
"value": "?search="
},
{
"operation": "search",
"time": "10.298",
"value": "Diana Dell"
},
{
"operation": "Now on",
"time": "10.989",
"value": "Diana Dell"
},
{
"operation": "Highlighting",
"time": "17.492",
"value": "Diana Dell_sentence_1"
},
{
"operation": "hyperlink",
"time": "21.947",
"value": "Grozny"
},
{
"operation": "Now on",
"time": "22.916",
"value": "Grozny"
},
{
"operation": "Highlighting",
"time": "27.482",
"value": "Grozny_sentence_2"
},
{
"operation": "back-button-clicked",
"time": "54.395",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "54.466",
"value": "Diana Dell"
},
{
"operation": "hyperlink",
"time": "62.02",
"value": "Grozny"
},
{
"operation": "Now on",
"time": "62.724",
"value": "Grozny"
},
{
"operation": "Highlighting deleted",
"time": "66.7",
"value": "Grozny_sentence_2"
},
{
"operation": "Highlighting",
"time": "67.42",
"value": "Grozny_sentence_1"
},
{
"operation": "Highlighting",
"time": "68.331",
"value": "Grozny_sentence_2"
},
{
"operation": "finish",
"time": "80.257",
"value": "finish"
}
]
}
|
85983
|
Tinchy Stryder third album Third Strike was released on Monday, 15 November 2010 and a year and a day later he released his second studio album.
|
{
"evidence": [
"Tinchy Stryder_sentence_38",
"Catch 22 (Tinchy Stryder album)_sentence_0",
"Catch 22 (Tinchy Stryder album)_sentence_2"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nTinchy Stryder third album Third Strike was released on Monday, 15 November 2010 and a year and a day later he released his second studio album.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Tinchy Stryder_sentence_38",
"Catch 22 (Tinchy Stryder album)_sentence_0",
"Catch 22 (Tinchy Stryder album)_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6725,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.339",
"value": "undefined"
},
{
"operation": "search",
"time": "9.481",
"value": "Tinchy Stryder"
},
{
"operation": "Now on",
"time": "10.953",
"value": "Tinchy Stryder"
},
{
"operation": "hyperlink",
"time": "23.824",
"value": "Third Strike"
},
{
"operation": "Now on",
"time": "24.382",
"value": "Third Strike (album)"
},
{
"operation": "Highlighting",
"time": "28.703",
"value": "Third Strike (album)_sentence_0"
},
{
"operation": "Highlighting",
"time": "28.705",
"value": "Third Strike (album)_title"
},
{
"operation": "Highlighting",
"time": "35.377",
"value": "Third Strike (album)_sentence_1"
},
{
"operation": "Page search",
"time": "43.684",
"value": "monday"
},
{
"operation": "back-button-clicked",
"time": "47.136",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "47.218",
"value": "Tinchy Stryder"
},
{
"operation": "Page search",
"time": "52.381",
"value": "monday"
},
{
"operation": "Highlighting",
"time": "61.992",
"value": "Tinchy Stryder_sentence_38"
},
{
"operation": "hyperlink",
"time": "302.916",
"value": "Catch 22"
},
{
"operation": "Now on",
"time": "303.707",
"value": "Catch 22 (Tinchy Stryder album)"
},
{
"operation": "Highlighting",
"time": "308.219",
"value": "Catch 22 (Tinchy Stryder album)_sentence_0"
},
{
"operation": "Highlighting",
"time": "309.39",
"value": "Catch 22 (Tinchy Stryder album)_sentence_2"
},
{
"operation": "finish",
"time": "330.839",
"value": "finish"
}
]
}
|
exfever_854
|
"Take a Bow" is a song by Madonna from Bedtime Stories, which is the successor to her fifth studio album that was released simultaneously with her 1982 book, with photography by Steven Meisel Studio and Fabien Baron.
|
{
"evidence": [
"Take a Bow (Madonna song)",
"Erotica (Madonna album)",
"Sex (book)"
],
"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\"Take a Bow\" is a song by Madonna from Bedtime Stories, which is the successor to her fifth studio album that was released simultaneously with her 1982 book, with photography by Steven Meisel Studio and Fabien Baron.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Take a Bow (Madonna song)",
"Erotica (Madonna album)",
"Sex (book)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6726,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1895
|
Julia Scarlett Elizabeth Louis-Dreyfus is one of the most award-winning actresses in American television history, tying an American actress and comedienne whose career spanned more than seven decades for the most acting wins.
|
{
"evidence": [
"Julia Louis-Dreyfus",
"Cloris Leachman"
],
"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:\nJulia Scarlett Elizabeth Louis-Dreyfus is one of the most award-winning actresses in American television history, tying an American actress and comedienne whose career spanned more than seven decades for the most acting wins.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Julia Louis-Dreyfus",
"Cloris Leachman"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6727,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
28729
|
In the 1990s, 16 top players left the BDO and formed the Professional Darts Corporation after popularity for the sport hit a low.
|
{
"evidence": [
"Professional Darts Corporation_sentence_7",
"Professional Darts Corporation_sentence_8"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nIn the 1990s, 16 top players left the BDO and formed the Professional Darts Corporation after popularity for the sport hit a low.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Professional Darts Corporation_sentence_7",
"Professional Darts Corporation_sentence_8"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6728,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.039",
"value": "?search="
},
{
"operation": "search",
"time": "16.997",
"value": "Contains...Split in darts"
},
{
"operation": "Now on",
"time": "18.497",
"value": "undefined"
},
{
"operation": "search",
"time": "32.81",
"value": "Professional Darts Corporation"
},
{
"operation": "Now on",
"time": "35.347",
"value": "Professional Darts Corporation"
},
{
"operation": "Page search",
"time": "46.619",
"value": "bdo"
},
{
"operation": "Highlighting",
"time": "49.735",
"value": "Professional Darts Corporation_sentence_8"
},
{
"operation": "Highlighting deleted",
"time": "54.386",
"value": "Professional Darts Corporation_sentence_8"
},
{
"operation": "Highlighting",
"time": "59.6",
"value": "Professional Darts Corporation_sentence_7"
},
{
"operation": "Highlighting",
"time": "67.322",
"value": "Professional Darts Corporation_sentence_8"
},
{
"operation": "finish",
"time": "132.195",
"value": "finish"
}
]
}
|
exfever_8874
|
"The Long Run" is a song recorded by the Eagles which was written by an American musician, singer, songwriter, and record producer who was the drummer and co-lead vocalist for the Eagles from 1971 until the band broke up in 1980.
|
{
"evidence": [
"The Long Run (song)",
"Don Henley"
],
"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:\n\"The Long Run\" is a song recorded by the Eagles which was written by an American musician, singer, songwriter, and record producer who was the drummer and co-lead vocalist for the Eagles from 1971 until the band broke up in 1980.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Long Run (song)",
"Don Henley"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6729,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1349
|
The Enough Project is a non-profit organization that conducts research in several conflict areas in Egypt that were controlled by a Christian extremist and a terrorist organization founded in 1987.
|
{
"evidence": [
"Enough Project",
"Lord's Resistance Army"
],
"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 Enough Project is a non-profit organization that conducts research in several conflict areas in Egypt that were controlled by a Christian extremist and a terrorist organization founded in 1987.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Enough Project",
"Lord's Resistance Army"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6730,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4001
|
Bengali is the official language of the states, including the valley located in the southern region of the Indian state of Assam.
|
{
"evidence": [
"Bengali language",
"Barak Valley"
],
"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:\nBengali is the official language of the states, including the valley located in the southern region of the Indian state of Assam.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Bengali language",
"Barak Valley"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6731,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
84700
|
Pioneer and Endicott Buildings was made by architect Cass Gilbert that has Renaissance style.
|
{
"evidence": [
"Pioneer and Endicott Buildings_sentence_2"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nPioneer and Endicott Buildings was made by architect Cass Gilbert that has Renaissance style.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Pioneer and Endicott Buildings_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6732,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.433",
"value": "?search="
},
{
"operation": "search",
"time": "30.822",
"value": "Pioneer and Endicott Buildings"
},
{
"operation": "Now on",
"time": "32.364",
"value": "Pioneer and Endicott Buildings"
},
{
"operation": "Highlighting",
"time": "36.644",
"value": "Pioneer and Endicott Buildings_cell_0_3_1"
},
{
"operation": "Highlighting",
"time": "36.648",
"value": "Pioneer and Endicott Buildings_header_cell_0_3_0"
},
{
"operation": "Highlighting",
"time": "37.798",
"value": "Pioneer and Endicott Buildings_cell_0_4_1"
},
{
"operation": "Highlighting",
"time": "37.801",
"value": "Pioneer and Endicott Buildings_header_cell_0_4_0"
},
{
"operation": "Highlighting",
"time": "46.733",
"value": "Pioneer and Endicott Buildings_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "48.45",
"value": "Pioneer and Endicott Buildings_cell_0_4_1"
},
{
"operation": "Highlighting deleted",
"time": "48.454",
"value": "Pioneer and Endicott Buildings_header_cell_0_4_0"
},
{
"operation": "Highlighting deleted",
"time": "49.142",
"value": "Pioneer and Endicott Buildings_cell_0_3_1"
},
{
"operation": "Highlighting deleted",
"time": "49.146",
"value": "Pioneer and Endicott Buildings_header_cell_0_3_0"
},
{
"operation": "finish",
"time": "61.527",
"value": "finish"
}
]
}
|
88665
|
Berkeley cardinals are certain large cardinals suggested by Hugh Woodin, but their existence is inconsistent with the axiom of countable choice.
|
{
"evidence": [
"Berkeley cardinal_sentence_0",
"Berkeley cardinal_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:\nBerkeley cardinals are certain large cardinals suggested by Hugh Woodin, but their existence is inconsistent with the axiom of countable choice.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Berkeley cardinal_sentence_0",
"Berkeley cardinal_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6733,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.131",
"value": "?search="
},
{
"operation": "search",
"time": "5.922",
"value": "Berkeley cardinals"
},
{
"operation": "Now on",
"time": "7.048",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "9.826",
"value": "Berkeley cardinal"
},
{
"operation": "Now on",
"time": "10.353",
"value": "Berkeley cardinal"
},
{
"operation": "Highlighting",
"time": "14.274",
"value": "Berkeley cardinal_sentence_0"
},
{
"operation": "Highlighting",
"time": "20.517",
"value": "Berkeley cardinal_sentence_2"
},
{
"operation": "finish",
"time": "27.023",
"value": "finish"
}
]
}
|
29748
|
Carbonaceous chondrite only has one subtype or group classified within it.
|
{
"evidence": [
"Carbonaceous chondrite_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:\nCarbonaceous chondrite only has one subtype or group classified within it.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Carbonaceous chondrite_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6734,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.794",
"value": "?search="
},
{
"operation": "search",
"time": "24.396",
"value": "Carbonaceous chondrite"
},
{
"operation": "Now on",
"time": "25.44",
"value": "Carbonaceous chondrite"
},
{
"operation": "Highlighting",
"time": "72.534",
"value": "Carbonaceous chondrite_sentence_0"
},
{
"operation": "finish",
"time": "136.106",
"value": "finish"
}
]
}
|
75503
|
I Like How It Feels is a song by Spanish recording artist Enrique Iglesias who is a singer, songwriter, actor, and record producer who is know as the King of Latin Pop who began his career during the mid-1990s..
|
{
"evidence": [
"I Like How It Feels_sentence_0",
"Enrique Iglesias_sentence_3",
"Enrique Iglesias_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:\nI Like How It Feels is a song by Spanish recording artist Enrique Iglesias who is a singer, songwriter, actor, and record producer who is know as the King of Latin Pop who began his career during the mid-1990s..\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"I Like How It Feels_sentence_0",
"Enrique Iglesias_sentence_3",
"Enrique Iglesias_sentence_4"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6735,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.928",
"value": "?search="
},
{
"operation": "search",
"time": "26.78",
"value": "I Like How It Feels"
},
{
"operation": "Now on",
"time": "28.335",
"value": "I Like How It Feels"
},
{
"operation": "Highlighting",
"time": "31.351",
"value": "I Like How It Feels_sentence_0"
},
{
"operation": "hyperlink",
"time": "33.894",
"value": "Enrique Iglesias"
},
{
"operation": "Now on",
"time": "35.26",
"value": "Enrique Iglesias"
},
{
"operation": "Highlighting",
"time": "39.081",
"value": "Enrique Iglesias_sentence_3"
},
{
"operation": "Highlighting",
"time": "47.619",
"value": "Enrique Iglesias_sentence_4"
},
{
"operation": "back-button-clicked",
"time": "55.976",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "56.128",
"value": "I Like How It Feels"
},
{
"operation": "finish",
"time": "65.582",
"value": "finish"
}
]
}
|
84526
|
Green Day's "Basket Case" was nominated for more than nine awards in 1995.
|
{
"evidence": [
"Basket Case (song)_sentence_28"
],
"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:\nGreen Day's \"Basket Case\" was nominated for more than nine awards in 1995.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Basket Case (song)_sentence_28"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6736,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.303",
"value": "?search="
},
{
"operation": "search",
"time": "9.303",
"value": "Basket Case"
},
{
"operation": "Now on",
"time": "9.948",
"value": "Basket Case"
},
{
"operation": "hyperlink",
"time": "14.401",
"value": "\"Basket Case\" (song)"
},
{
"operation": "Now on",
"time": "15.072",
"value": "Basket Case (song)"
},
{
"operation": "Highlighting",
"time": "101.746",
"value": "Basket Case (song)_sentence_28"
},
{
"operation": "finish",
"time": "128.169",
"value": "finish"
}
]
}
|
47016
|
Open access to scholarly communication in Belgium was slowed after 2007, until the Brussels Declaration for open access was officially signed in 2012.
|
{
"evidence": [
"Open access in Belgium_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:\nOpen access to scholarly communication in Belgium was slowed after 2007, until the Brussels Declaration for open access was officially signed in 2012.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Open access in Belgium_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6737,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.746",
"value": "?search="
},
{
"operation": "search",
"time": "19.71",
"value": "Brussels Declaration for open access"
},
{
"operation": "Now on",
"time": "21.833",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "24.557",
"value": "Open access in Belgium"
},
{
"operation": "Now on",
"time": "25.143",
"value": "Open access in Belgium"
},
{
"operation": "Highlighting",
"time": "28.739",
"value": "Open access in Belgium_sentence_0"
},
{
"operation": "finish",
"time": "42.911",
"value": "finish"
}
]
}
|
53213
|
Nangma is a genre of Tibetan dance music whose name is derived from a Persian word for dance.
|
{
"evidence": [
"Nangma_sentence_0",
"Nangma_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:\nNangma is a genre of Tibetan dance music whose name is derived from a Persian word for dance.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Nangma_sentence_0",
"Nangma_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6738,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.516",
"value": "?search="
},
{
"operation": "search",
"time": "15.28",
"value": "Nangma"
},
{
"operation": "Now on",
"time": "17.023",
"value": "Nangma"
},
{
"operation": "Highlighting",
"time": "20.75",
"value": "Nangma_sentence_0"
},
{
"operation": "Highlighting",
"time": "24.84",
"value": "Nangma_sentence_1"
},
{
"operation": "finish",
"time": "50.144",
"value": "finish"
}
]
}
|
exfever_6981
|
Cory Allan Michael Monteith (; May 11, 1982 – July 13, 2013) was a Canadian actor and musician best known for his role as a fictional character from the HBO musical comedy-drama series Glee.
|
{
"evidence": [
"Cory Monteith",
"Finn Hudson"
],
"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:\nCory Allan Michael Monteith (; May 11, 1982 – July 13, 2013) was a Canadian actor and musician best known for his role as a fictional character from the HBO musical comedy-drama series Glee.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Cory Monteith",
"Finn Hudson"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6739,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
60130
|
One manual depicted Walt Kelly's friend Joseph McCarthy as a caveman in the same period that saw the creation of Kelly's most famous character, Pogo, a daily comic strip that was created by Kelly and syndicated to American newspapers from 1948 until 1975.
|
{
"evidence": [
"Walt Kelly_sentence_22",
"Walt Kelly_sentence_37"
],
"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:\nOne manual depicted Walt Kelly's friend Joseph McCarthy as a caveman in the same period that saw the creation of Kelly's most famous character, Pogo, a daily comic strip that was created by Kelly and syndicated to American newspapers from 1948 until 1975.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Walt Kelly_sentence_22",
"Walt Kelly_sentence_37"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6740,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.213",
"value": "?search="
},
{
"operation": "search",
"time": "30.39",
"value": "Walt Kelly"
},
{
"operation": "Now on",
"time": "31.528",
"value": "Walt Kelly"
},
{
"operation": "Page search",
"time": "37.61",
"value": "one manual"
},
{
"operation": "Highlighting",
"time": "43.445",
"value": "Walt Kelly_sentence_22"
},
{
"operation": "Page search",
"time": "65.132",
"value": "joseph"
},
{
"operation": "Highlighting",
"time": "94.869",
"value": "Walt Kelly_sentence_37"
},
{
"operation": "Now on",
"time": "193.651",
"value": "Walt Kelly"
},
{
"operation": "finish",
"time": "208.093",
"value": "finish"
}
]
}
|
exfever_6959
|
Rope is a 1948 American crime film directed by Alfred Hitchcock whose genre combines the thriller and psychological fiction genres.
|
{
"evidence": [
"Rope (film)",
"Psychological thriller"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nRope is a 1948 American crime film directed by Alfred Hitchcock whose genre combines the thriller and psychological fiction genres.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Rope (film)",
"Psychological thriller"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6741,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
94167
|
Quinalow was affected by the 2010–2011 Queensland floods, including the Quinalow Library opened in March 1984.
|
{
"evidence": [
"Quinalow_sentence_7",
"Quinalow_sentence_8",
"Quinalow_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:\nQuinalow was affected by the 2010–2011 Queensland floods, including the Quinalow Library opened in March 1984.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Quinalow_sentence_7",
"Quinalow_sentence_8",
"Quinalow_sentence_6"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6742,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.425",
"value": "?search="
},
{
"operation": "search",
"time": "8.966",
"value": "Quinalow"
},
{
"operation": "Now on",
"time": "9.675",
"value": "Quinalow"
},
{
"operation": "Highlighting",
"time": "26.214",
"value": "Quinalow_sentence_7"
},
{
"operation": "Highlighting",
"time": "33.102",
"value": "Quinalow_sentence_8"
},
{
"operation": "Highlighting",
"time": "53.582",
"value": "Quinalow_sentence_6"
},
{
"operation": "search",
"time": "63.925",
"value": "Quinalow Library"
},
{
"operation": "Now on",
"time": "64.588",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "68.502",
"value": "Library Quinalow_section_2"
},
{
"operation": "Now on",
"time": "68.911",
"value": "Quinalow#Library Quinalow section 2"
},
{
"operation": "hyperlink",
"time": "74.286",
"value": "Toowoomba Regional Council"
},
{
"operation": "Now on",
"time": "75.157",
"value": "Toowoomba Region"
},
{
"operation": "Page search",
"time": "92.984",
"value": "library"
},
{
"operation": "hyperlink",
"time": "98.918",
"value": "7 Toowoomba Regional Library Services Toowoomba Region_section_9"
},
{
"operation": "hyperlink",
"time": "104.31",
"value": "Quinalow"
},
{
"operation": "Now on",
"time": "104.825",
"value": "Quinalow"
},
{
"operation": "finish",
"time": "134.414",
"value": "finish"
}
]
}
|
exfever_3008
|
Heather Joan Graham’s second starring role in a 1988 American teen comedy film written by Neil Tolkin.
|
{
"evidence": [
"Heather Graham",
"License to Drive"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nHeather Joan Graham’s second starring role in a 1988 American teen comedy film written by Neil Tolkin.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Heather Graham",
"License to Drive"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6743,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8006
|
Elementary is an American procedural drama television series that starred a theatre actor who achieved early success for his portrayal of Simon "Sick Boy" Williamson in the dark comedy-drama film Trainspotting.
|
{
"evidence": [
"Elementary (TV series)",
"Jonny Lee Miller"
],
"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:\nElementary is an American procedural drama television series that starred a theatre actor who achieved early success for his portrayal of Simon \"Sick Boy\" Williamson in the dark comedy-drama film Trainspotting.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Elementary (TV series)",
"Jonny Lee Miller"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6744,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
39658
|
The highest recorded Mean Absolute Error, a measure of errors between paired observations expressing the same phenomenon, in the Kalawa Polling Division was in 1988 and 1982.
|
{
"evidence": [
"Mean absolute error_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 highest recorded Mean Absolute Error, a measure of errors between paired observations expressing the same phenomenon, in the Kalawa Polling Division was in 1988 and 1982.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Mean absolute error_sentence_1"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6745,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.678",
"value": "?search="
},
{
"operation": "search",
"time": "11.511",
"value": "Mean absolute error"
},
{
"operation": "Now on",
"time": "12.483",
"value": "Mean absolute error"
},
{
"operation": "Page search",
"time": "17.854",
"value": "highest"
},
{
"operation": "Highlighting",
"time": "41.18",
"value": "Mean absolute error_sentence_1"
},
{
"operation": "search",
"time": "61.686",
"value": "Contains...Kalawa polling divison"
},
{
"operation": "Now on",
"time": "62.528",
"value": "undefined"
},
{
"operation": "search",
"time": "108.868",
"value": "Mbooni Constituency"
},
{
"operation": "Now on",
"time": "109.593",
"value": "Mbooni Constituency"
},
{
"operation": "Page search",
"time": "119.303",
"value": "Kalawa"
},
{
"operation": "Highlighting",
"time": "124.055",
"value": "Mbooni Constituency_cell_1_6_1"
},
{
"operation": "Highlighting",
"time": "124.057",
"value": "Mbooni Constituency_title"
},
{
"operation": "Highlighting",
"time": "124.058",
"value": "Mbooni Constituency_section_1"
},
{
"operation": "Highlighting",
"time": "124.059",
"value": "Mbooni Constituency_header_cell_1_2_1"
},
{
"operation": "Highlighting deleted",
"time": "124.251",
"value": "Mbooni Constituency_cell_1_6_1"
},
{
"operation": "Highlighting deleted",
"time": "124.253",
"value": "Mbooni Constituency_title"
},
{
"operation": "Highlighting deleted",
"time": "124.255",
"value": "Mbooni Constituency_section_1"
},
{
"operation": "Highlighting deleted",
"time": "124.257",
"value": "Mbooni Constituency_header_cell_1_2_1"
},
{
"operation": "Highlighting",
"time": "125.56",
"value": "Mbooni Constituency_cell_1_5_0"
},
{
"operation": "Highlighting",
"time": "125.56",
"value": "Mbooni Constituency_title"
},
{
"operation": "Highlighting",
"time": "125.56",
"value": "Mbooni Constituency_section_1"
},
{
"operation": "Highlighting",
"time": "125.561",
"value": "Mbooni Constituency_header_cell_1_2_0"
},
{
"operation": "Highlighting deleted",
"time": "126.191",
"value": "Mbooni Constituency_cell_1_5_0"
},
{
"operation": "Highlighting deleted",
"time": "126.193",
"value": "Mbooni Constituency_title"
},
{
"operation": "Highlighting deleted",
"time": "126.194",
"value": "Mbooni Constituency_section_1"
},
{
"operation": "Highlighting deleted",
"time": "126.195",
"value": "Mbooni Constituency_header_cell_1_2_0"
},
{
"operation": "search",
"time": "132.458",
"value": "Kalawa"
},
{
"operation": "Now on",
"time": "133.312",
"value": "Kalawa"
},
{
"operation": "search",
"time": "155.565",
"value": "Kalawa Jazmee Records"
},
{
"operation": "Now on",
"time": "156.247",
"value": "Kalawa Jazmee Records"
},
{
"operation": "Page search",
"time": "162.307",
"value": "division"
},
{
"operation": "search",
"time": "184.358",
"value": "Contains...Kalawa Polling Division"
},
{
"operation": "Now on",
"time": "185.22",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "190.766",
"value": "kagawa polling division"
},
{
"operation": "Now on",
"time": "191.33",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "193.304",
"value": "2020 in Japan"
},
{
"operation": "Now on",
"time": "194.559",
"value": "2020 in Japan"
},
{
"operation": "Page search",
"time": "198.811",
"value": "kalawa"
},
{
"operation": "finish",
"time": "279.274",
"value": "finish"
}
]
}
|
exfever_4342
|
One Direction is an English-Irish pop boy band who signed with Simon Cowell's record label, a defunct division of Syco Entertainment that oversaw releases of the Irish boy band formed in Sligo in 1999.
|
{
"evidence": [
"One Direction",
"Syco Music",
"Westlife"
],
"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 Direction is an English-Irish pop boy band who signed with Simon Cowell's record label, a defunct division of Syco Entertainment that oversaw releases of the Irish boy band formed in Sligo in 1999.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"One Direction",
"Syco Music",
"Westlife"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6746,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3550
|
IBM was the first computer company to manufacture modern laptops.
|
{
"evidence": [
"IBM",
"Armonk, New York"
],
"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:\nIBM was the first computer company to manufacture modern laptops.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"IBM",
"Armonk, New York"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6747,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
21808
|
The Chats released "Pub Feed", the band toured the United States but never performed in the UK.
|
{
"evidence": [
"The Chats_sentence_28"
],
"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 Chats released \"Pub Feed\", the band toured the United States but never performed in the UK.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"The Chats_sentence_28"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6748,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.617",
"value": "?search="
},
{
"operation": "search",
"time": "7.105",
"value": "Pub Feed"
},
{
"operation": "Now on",
"time": "7.944",
"value": "The Chats"
},
{
"operation": "Highlighting",
"time": "10.556",
"value": "The Chats_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "20.155",
"value": "The Chats_sentence_0"
},
{
"operation": "Page search",
"time": "36.539",
"value": "United States"
},
{
"operation": "Highlighting",
"time": "44.092",
"value": "The Chats_sentence_28"
},
{
"operation": "finish",
"time": "51.325",
"value": "finish"
}
]
}
|
75549
|
Alexei Korzukhin (1835—1894) was a German genre painter who was born into a family of gold prospectors, entered the Imperial Academy of Fine Arts in 1858 and witness to the extremely bloody assassination of Tsar Alexander II whcih resulted in severe nervous shock .
|
{
"evidence": [
"Alexei Korzukhin_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:\nAlexei Korzukhin (1835—1894) was a German genre painter who was born into a family of gold prospectors, entered the Imperial Academy of Fine Arts in 1858 and witness to the extremely bloody assassination of Tsar Alexander II whcih resulted in severe nervous shock .\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Alexei Korzukhin_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6749,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.703",
"value": "?search="
},
{
"operation": "search",
"time": "19.406",
"value": "Alexei Korzukhin"
},
{
"operation": "Now on",
"time": "20.446",
"value": "Alexei Korzukhin"
},
{
"operation": "Highlighting",
"time": "189.181",
"value": "Alexei Korzukhin_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "189.757",
"value": "Alexei Korzukhin_sentence_0"
},
{
"operation": "Highlighting",
"time": "190.501",
"value": "Alexei Korzukhin_sentence_0"
},
{
"operation": "finish",
"time": "204.869",
"value": "finish"
}
]
}
|
76704
|
Basic LRU (least recently used) maintains an ordered list, new entries are added at the top of the list and cache hits move to the top.
|
{
"evidence": [
"Adaptive replacement cache_sentence_4",
"Adaptive replacement cache_sentence_5",
"Adaptive replacement cache_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:\nBasic LRU (least recently used) maintains an ordered list, new entries are added at the top of the list and cache hits move to the top.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Adaptive replacement cache_sentence_4",
"Adaptive replacement cache_sentence_5",
"Adaptive replacement cache_sentence_6"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6750,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.751",
"value": "?search="
},
{
"operation": "search",
"time": "18.701",
"value": "LRU"
},
{
"operation": "Now on",
"time": "19.393",
"value": "LRU"
},
{
"operation": "Highlighting",
"time": "26.423",
"value": "LRU_item_0_2"
},
{
"operation": "Highlighting",
"time": "26.424",
"value": "LRU_title"
},
{
"operation": "hyperlink",
"time": "29.439",
"value": "Least Recently Used"
},
{
"operation": "Now on",
"time": "30.529",
"value": "Cache replacement policies#Least recently used .28LRU.29"
},
{
"operation": "Page search",
"time": "43.143",
"value": "ordered list"
},
{
"operation": "Highlighting",
"time": "76.44",
"value": "Cache replacement policies_sentence_135"
},
{
"operation": "Highlighting",
"time": "76.441",
"value": "Cache replacement policies_title"
},
{
"operation": "Highlighting",
"time": "76.442",
"value": "Cache replacement policies_section_1"
},
{
"operation": "Highlighting",
"time": "76.442",
"value": "Cache replacement policies_section_18"
},
{
"operation": "search",
"time": "83.799",
"value": "Adaptive replacement cache"
},
{
"operation": "Now on",
"time": "84.684",
"value": "Adaptive replacement cache"
},
{
"operation": "Highlighting",
"time": "94.799",
"value": "Adaptive replacement cache_sentence_4"
},
{
"operation": "Highlighting",
"time": "108.584",
"value": "Adaptive replacement cache_sentence_5"
},
{
"operation": "Highlighting",
"time": "112.049",
"value": "Adaptive replacement cache_sentence_6"
},
{
"operation": "finish",
"time": "121.508",
"value": "finish"
}
]
}
|
37570
|
The Campbell's mona monkey, qualified as a near-threatened species, was named for Henry Dundas Campbell, Sierra Lione's Lieutenant-Governor in 1832 (and later its Governor).
|
{
"evidence": [
"Campbell's mona monkey_sentence_1",
"Henry Dundas Campbell_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:\nThe Campbell's mona monkey, qualified as a near-threatened species, was named for Henry Dundas Campbell, Sierra Lione's Lieutenant-Governor in 1832 (and later its Governor).\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Campbell's mona monkey_sentence_1",
"Henry Dundas Campbell_sentence_6"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6751,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.13",
"value": "?search="
},
{
"operation": "search",
"time": "12.935",
"value": "Campbell's mona monkey"
},
{
"operation": "Now on",
"time": "14.191",
"value": "Campbell's mona monkey"
},
{
"operation": "Highlighting",
"time": "34.204",
"value": "Campbell's mona monkey_sentence_3"
},
{
"operation": "Highlighting",
"time": "44.846",
"value": "Campbell's mona monkey_sentence_1"
},
{
"operation": "hyperlink",
"time": "54.094",
"value": "Henry Dundas Campbell"
},
{
"operation": "Now on",
"time": "54.784",
"value": "Henry Dundas Campbell"
},
{
"operation": "Highlighting",
"time": "63.274",
"value": "Henry Dundas Campbell_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "66.941",
"value": "Henry Dundas Campbell_sentence_0"
},
{
"operation": "Highlighting",
"time": "75.593",
"value": "Henry Dundas Campbell_sentence_6"
},
{
"operation": "back-button-clicked",
"time": "87.116",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "87.346",
"value": "Campbell's mona monkey"
},
{
"operation": "Highlighting deleted",
"time": "93.966",
"value": "Campbell's mona monkey_sentence_3"
},
{
"operation": "finish",
"time": "101.829",
"value": "finish"
}
]
}
|
exfever_5752
|
21 Jump Street is a 2012 American buddy cop action comedy series written by an American actor, filmmaker, and comedian who is known for his comedic roles in films including Superbad.
|
{
"evidence": [
"21 Jump Street (film)",
"Jonah Hill"
],
"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:\n21 Jump Street is a 2012 American buddy cop action comedy series written by an American actor, filmmaker, and comedian who is known for his comedic roles in films including Superbad.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"21 Jump Street (film)",
"Jonah Hill"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6752,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
52167
|
Dundee United and St. Johnstone competed in the 1991 Scottish Cup Final.
|
{
"evidence": [
"1991 Scottish Cup Final_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:\nDundee United and St. Johnstone competed in the 1991 Scottish Cup Final.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"1991 Scottish Cup Final_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6753,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.077",
"value": "?search="
},
{
"operation": "search",
"time": "25.263",
"value": "1991 Scottish Cup Final"
},
{
"operation": "Now on",
"time": "26.173",
"value": "1991 Scottish Cup Final"
},
{
"operation": "Highlighting",
"time": "42.935",
"value": "1991 Scottish Cup Final_sentence_1"
},
{
"operation": "Page search",
"time": "51.032",
"value": "johnstone"
},
{
"operation": "page-search-reset",
"time": "75.318",
"value": "page-search-reset"
},
{
"operation": "finish",
"time": "185.837",
"value": "finish"
}
]
}
|
exfever_1635
|
The Who are a rock band formed in the capital and largest city of a country that is part of the United States.
|
{
"evidence": [
"The Who",
"London",
"England"
],
"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 Who are a rock band formed in the capital and largest city of a country that is part of the United States.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Who",
"London",
"England"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6754,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5999
|
Penny Dreadful is a comedy-drama television series created for an American premium television network owned by Paramount Media Networks and Sky by John Logan.
|
{
"evidence": [
"Penny Dreadful (TV series)",
"Showtime (TV network)"
],
"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:\nPenny Dreadful is a comedy-drama television series created for an American premium television network owned by Paramount Media Networks and Sky by John Logan.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Penny Dreadful (TV series)",
"Showtime (TV network)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6755,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
91648
|
Alexander Osterwalder was a journalist for BILANZ for one year and founded BusinessModelDesign.com five years later.
|
{
"evidence": [
"Alexander Osterwalder_sentence_4",
"Alexander Osterwalder_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:\nAlexander Osterwalder was a journalist for BILANZ for one year and founded BusinessModelDesign.com five years later.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Alexander Osterwalder_sentence_4",
"Alexander Osterwalder_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6756,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.494",
"value": "?search="
},
{
"operation": "search",
"time": "9.543",
"value": "Alexander Osterwalder"
},
{
"operation": "Now on",
"time": "10.876",
"value": "Alexander Osterwalder"
},
{
"operation": "Highlighting",
"time": "14.155",
"value": "Alexander Osterwalder_sentence_4"
},
{
"operation": "Highlighting",
"time": "19.206",
"value": "Alexander Osterwalder_sentence_5"
},
{
"operation": "finish",
"time": "30.567",
"value": "finish"
}
]
}
|
70061
|
Félix Baloy, a vocalist with many groups, including Chappotin y Sus Estrellas, Revé, and Adalberto Alvarez y su Son was one of the original members of the Afro-Cuban All Stars and sang on the compilation album Café Cubano, produced through Putumayo World Music.
|
{
"evidence": [
"Félix Baloy_sentence_2",
"Félix Baloy_sentence_3",
"Félix Baloy_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:\nFélix Baloy, a vocalist with many groups, including Chappotin y Sus Estrellas, Revé, and Adalberto Alvarez y su Son was one of the original members of the Afro-Cuban All Stars and sang on the compilation album Café Cubano, produced through Putumayo World Music.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Félix Baloy_sentence_2",
"Félix Baloy_sentence_3",
"Félix Baloy_sentence_4"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6757,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.771",
"value": "?search="
},
{
"operation": "search",
"time": "13.075",
"value": "Félix Baloy"
},
{
"operation": "Now on",
"time": "13.604",
"value": "Félix Baloy"
},
{
"operation": "Highlighting",
"time": "24.209",
"value": "Félix Baloy_sentence_2"
},
{
"operation": "Highlighting",
"time": "36.273",
"value": "Félix Baloy_sentence_3"
},
{
"operation": "Highlighting",
"time": "40.617",
"value": "Félix Baloy_sentence_4"
},
{
"operation": "finish",
"time": "215.096",
"value": "finish"
}
]
}
|
45901
|
The writer of Because I Got High sums up that as a result of his constant reliance on cannabis, he lost custody of his children, lost his wife, and ended up homeless and "sleeping on the sidewalk" and admits that he is singing the whole thing wrong, because he is high.
|
{
"evidence": [
"Because I Got High_sentence_6",
"Because I Got High_sentence_7"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe writer of Because I Got High sums up that as a result of his constant reliance on cannabis, he lost custody of his children, lost his wife, and ended up homeless and \"sleeping on the sidewalk\" and admits that he is singing the whole thing wrong, because he is high.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Because I Got High_sentence_6",
"Because I Got High_sentence_7"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6758,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.579",
"value": "?search="
},
{
"operation": "search",
"time": "18.092",
"value": "Because I Got High"
},
{
"operation": "Now on",
"time": "19.655",
"value": "Because I Got High"
},
{
"operation": "Highlighting",
"time": "51.153",
"value": "Because I Got High_sentence_5"
},
{
"operation": "Highlighting",
"time": "54.921",
"value": "Because I Got High_sentence_6"
},
{
"operation": "Highlighting deleted",
"time": "55.639",
"value": "Because I Got High_sentence_5"
},
{
"operation": "Highlighting",
"time": "67.115",
"value": "Because I Got High_sentence_7"
},
{
"operation": "finish",
"time": "95.041",
"value": "finish"
}
]
}
|
exfever_1656
|
Asceticism is the most widely practiced tradition in the world.
|
{
"evidence": [
"Asceticism",
"Jainism",
"Religion"
],
"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:\nAsceticism is the most widely practiced tradition in the world.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Asceticism",
"Jainism",
"Religion"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6759,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7254
|
Lisbon is the capital and the largest city in a country located in the Iberian Peninsula of Southwestern Europe.
|
{
"evidence": [
"Lisbon",
"Portugal"
],
"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:\nLisbon is the capital and the largest city in a country located in the Iberian Peninsula of Southwestern Europe.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Lisbon",
"Portugal"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6760,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
51621
|
One Touch of Venus is based on a novella called The Tinted Venus that was on Broadway but never made into a film.
|
{
"evidence": [
"One Touch of Venus_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:\nOne Touch of Venus is based on a novella called The Tinted Venus that was on Broadway but never made into a film.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"One Touch of Venus_sentence_9"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6761,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.753",
"value": "?search="
},
{
"operation": "search",
"time": "14.089",
"value": "One Touch of Venus"
},
{
"operation": "Now on",
"time": "15.818",
"value": "One Touch of Venus"
},
{
"operation": "hyperlink",
"time": "23.656",
"value": "2 Film version One Touch of Venus_section_1"
},
{
"operation": "Highlighting",
"time": "28.051",
"value": "One Touch of Venus_sentence_9"
},
{
"operation": "finish",
"time": "53.876",
"value": "finish"
}
]
}
|
exfever_6019
|
John Hartford was an American folk, country, and bluegrass composer who invented his own shuffle tap dance move, and a type of folk dance which the dancer's footwear is used percussively by striking the heel onto an amplified piece of plywood while he played and sang, that is also he official state dance of the 28th largest and 9th-most populous of the 50 United States.
|
{
"evidence": [
"John Hartford",
"Clogging",
"North Carolina"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJohn Hartford was an American folk, country, and bluegrass composer who invented his own shuffle tap dance move, and a type of folk dance which the dancer's footwear is used percussively by striking the heel onto an amplified piece of plywood while he played and sang, that is also he official state dance of the 28th largest and 9th-most populous of the 50 United States.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"John Hartford",
"Clogging",
"North Carolina"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6762,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1451
|
American Horror Story is an anthology horror television series starring an American actor who plays multiple roles on an German pay television channel.
|
{
"evidence": [
"American Horror Story",
"Evan Peters",
"FX (TV channel)"
],
"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:\nAmerican Horror Story is an anthology horror television series starring an American actor who plays multiple roles on an German pay television channel.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"American Horror Story",
"Evan Peters",
"FX (TV channel)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6763,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7470
|
Judy Greer first rose to prominence in the late 1990s to early 2000s, appearing in a romantic fantasy comedy film directed by an Australian filmmaker.
|
{
"evidence": [
"Judy Greer",
"What Women Want",
"Nancy Meyers"
],
"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:\nJudy Greer first rose to prominence in the late 1990s to early 2000s, appearing in a romantic fantasy comedy film directed by an Australian filmmaker.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Judy Greer",
"What Women Want",
"Nancy Meyers"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6764,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
59670
|
"Nothin's Gonna Stop Us" and "One Way Ticket" of The Darkness (which had 24 music videos and 29 singles) was released in the year 2000.
|
{
"evidence": [
"Nothing's Gonna Stop Us_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:\n\"Nothin's Gonna Stop Us\" and \"One Way Ticket\" of The Darkness (which had 24 music videos and 29 singles) was released in the year 2000.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Nothing's Gonna Stop Us_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6765,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.041",
"value": "?search="
},
{
"operation": "search",
"time": "19.969",
"value": "The Darkness"
},
{
"operation": "Now on",
"time": "21.781",
"value": "Darkness (disambiguation)"
},
{
"operation": "hyperlink",
"time": "35.773",
"value": "3 In music Darkness (disambiguation)_section_2"
},
{
"operation": "hyperlink",
"time": "42.492",
"value": "The Darkness (band)"
},
{
"operation": "Now on",
"time": "43.713",
"value": "The Darkness (band)"
},
{
"operation": "Page search",
"time": "57.265",
"value": "Nothin's Gonna Stop Us"
},
{
"operation": "search",
"time": "94.947",
"value": "The Darkness discography"
},
{
"operation": "Now on",
"time": "96.218",
"value": "The Darkness discography"
},
{
"operation": "Page search",
"time": "102.651",
"value": "Nothin's Gonna Stop Us"
},
{
"operation": "hyperlink",
"time": "133.216",
"value": "Nothin's Gonna Stop Us"
},
{
"operation": "Now on",
"time": "134.212",
"value": "Nothing's Gonna Stop Us"
},
{
"operation": "Highlighting",
"time": "159.18",
"value": "Nothing's Gonna Stop Us_sentence_1"
},
{
"operation": "finish",
"time": "226.78",
"value": "finish"
}
]
}
|
exfever_1519
|
The Monster was written by a team of songwriters, including an American songwriter who was raised in Lake Grove, New York.
|
{
"evidence": [
"The Monster (song)",
"Jon Bellion"
],
"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 Monster was written by a team of songwriters, including an American songwriter who was raised in Lake Grove, New York.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Monster (song)",
"Jon Bellion"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6766,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
40492
|
Tobias Brodtkorp Bernhoft was a Swedish engineer and director, and then married Mary Bush on October 05, 1951, but didn't have any children with her.
|
{
"evidence": [
"Tobias Bernhoft_sentence_0",
"Tobias Bernhoft_sentence_2"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nTobias Brodtkorp Bernhoft was a Swedish engineer and director, and then married Mary Bush on October 05, 1951, but didn't have any children with her.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Tobias Bernhoft_sentence_0",
"Tobias Bernhoft_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6767,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.441",
"value": "?search="
},
{
"operation": "search",
"time": "19.515",
"value": "Contains...Tobias Brodtkorp Bernhoft"
},
{
"operation": "Now on",
"time": "20.641",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "30.094",
"value": "Tobias Bernhoft"
},
{
"operation": "Now on",
"time": "31.727",
"value": "Tobias Bernhoft"
},
{
"operation": "Highlighting",
"time": "48.649",
"value": "Tobias Bernhoft_sentence_0"
},
{
"operation": "Highlighting",
"time": "52.585",
"value": "Tobias Bernhoft_sentence_2"
},
{
"operation": "finish",
"time": "74.739",
"value": "finish"
}
]
}
|
91973
|
2003 Atlanta Braves season marked the franchise's 38th season where they won their 9th consecutive division title and lost the 2003 Divisional Series
|
{
"evidence": [
"2003 Atlanta Braves season_sentence_0",
"2003 Atlanta Braves season_sentence_2",
"2003 Atlanta Braves season_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:\n2003 Atlanta Braves season marked the franchise's 38th season where they won their 9th consecutive division title and lost the 2003 Divisional Series\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"2003 Atlanta Braves season_sentence_0",
"2003 Atlanta Braves season_sentence_2",
"2003 Atlanta Braves season_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6768,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.164",
"value": "?search="
},
{
"operation": "search",
"time": "19.814",
"value": "2003 Atlanta Braves season"
},
{
"operation": "Now on",
"time": "21.454",
"value": "2003 Atlanta Braves season"
},
{
"operation": "Highlighting",
"time": "25.447",
"value": "2003 Atlanta Braves season_sentence_0"
},
{
"operation": "Highlighting",
"time": "473.54",
"value": "2003 Atlanta Braves season_sentence_2"
},
{
"operation": "Highlighting",
"time": "479.194",
"value": "2003 Atlanta Braves season_sentence_1"
},
{
"operation": "finish",
"time": "493.102",
"value": "finish"
}
]
}
|
43506
|
Oxalate oxidase belongs to the family of oxidoreductases, the systematic name of this enzyme class is calcium oxalate which is the primary constituent of the most common kind of kidney stones.
|
{
"evidence": [
"Oxalate oxidase_sentence_3"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nOxalate oxidase belongs to the family of oxidoreductases, the systematic name of this enzyme class is calcium oxalate which is the primary constituent of the most common kind of kidney stones.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Oxalate oxidase_sentence_3"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6769,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.124",
"value": "?search="
},
{
"operation": "search",
"time": "25.286",
"value": "Oxalate oxidase"
},
{
"operation": "Now on",
"time": "26.187",
"value": "Oxalate oxidase"
},
{
"operation": "Highlighting",
"time": "85.098",
"value": "Oxalate oxidase_sentence_2"
},
{
"operation": "Highlighting",
"time": "94.746",
"value": "Oxalate oxidase_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "110.346",
"value": "Oxalate oxidase_sentence_2"
},
{
"operation": "finish",
"time": "116.719",
"value": "finish"
}
]
}
|
exfever_2894
|
Southport has a population of 90,336 and is a large seaside town in a metropolitan and ceremonial county in North West England.
|
{
"evidence": [
"Southport",
"Merseyside"
],
"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:\nSouthport has a population of 90,336 and is a large seaside town in a metropolitan and ceremonial county in North West England.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Southport",
"Merseyside"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6770,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
26052
|
Zhakyp Kozhamberdi started playing on June 14, 2013 for Taraz and appeared in all of the games for that season.
|
{
"evidence": [
"Zhakyp Kozhamberdi_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:\nZhakyp Kozhamberdi started playing on June 14, 2013 for Taraz and appeared in all of the games for that season.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Zhakyp Kozhamberdi_sentence_1"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6771,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.625",
"value": "?search="
},
{
"operation": "search",
"time": "12.303",
"value": "Zhakyp Kozhamberdi"
},
{
"operation": "Now on",
"time": "13.395",
"value": "Zhakyp Kozhamberdi"
},
{
"operation": "Highlighting",
"time": "34.742",
"value": "Zhakyp Kozhamberdi_sentence_1"
},
{
"operation": "Page search",
"time": "60.016",
"value": "all"
},
{
"operation": "Page search",
"time": "60.194",
"value": "all"
},
{
"operation": "Page search",
"time": "60.437",
"value": "all"
},
{
"operation": "Page search",
"time": "60.652",
"value": "all"
},
{
"operation": "finish",
"time": "221.999",
"value": "finish"
}
]
}
|
exfever_2483
|
Vienna is the national capital and one of the states that can pass laws that stay within the limits of the constitution.
|
{
"evidence": [
"Vienna",
"States of Austria"
],
"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:\nVienna is the national capital and one of the states that can pass laws that stay within the limits of the constitution.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Vienna",
"States of Austria"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6772,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
23087
|
Sweden's Jonas Sandqvist made the most appearances (26) in the Allsvenkan league, a Swedish professional league for men's association football clubs, among the 2007 Malmö FF season squad members.
|
{
"evidence": [
"Jonas Sandqvist_sentence_0",
"Malmö FF_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:\nSweden's Jonas Sandqvist made the most appearances (26) in the Allsvenkan league, a Swedish professional league for men's association football clubs, among the 2007 Malmö FF season squad members.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Jonas Sandqvist_sentence_0",
"Malmö FF_sentence_3"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6773,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.361",
"value": "?search="
},
{
"operation": "search",
"time": "30.153",
"value": "Contains...Sweden's Jonas Sandqvist"
},
{
"operation": "Now on",
"time": "31.76",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "40.527",
"value": "Jonas Sandqvist"
},
{
"operation": "Now on",
"time": "41.239",
"value": "Jonas Sandqvist"
},
{
"operation": "Highlighting",
"time": "49.211",
"value": "Jonas Sandqvist_sentence_0"
},
{
"operation": "Now on",
"time": "129.217",
"value": "Jonas Sandqvist"
},
{
"operation": "hyperlink",
"time": "277.509",
"value": "Malmö FF"
},
{
"operation": "Now on",
"time": "279.648",
"value": "Malmö FF"
},
{
"operation": "Highlighting",
"time": "287.978",
"value": "Malmö FF_sentence_3"
},
{
"operation": "search",
"time": "442.658",
"value": "Allsvenkan league"
},
{
"operation": "Now on",
"time": "444.285",
"value": "undefined"
},
{
"operation": "search",
"time": "485.688",
"value": "Allsvenkan"
},
{
"operation": "Now on",
"time": "486.297",
"value": "undefined"
},
{
"operation": "search",
"time": "540.741",
"value": "2007 Allsvenka"
},
{
"operation": "Now on",
"time": "542.077",
"value": "undefined"
},
{
"operation": "search",
"time": "577.948",
"value": "Jonas Sandqvist"
},
{
"operation": "Now on",
"time": "579.076",
"value": "Jonas Sandqvist"
},
{
"operation": "Now on",
"time": "704.286",
"value": "Jonas Sandqvist"
},
{
"operation": "finish",
"time": "713.52",
"value": "finish"
}
]
}
|
66142
|
The Virtual particle's concept arises in the perturbation theory of quantum field theory, where the calculations are often performed using schematic representations known as Feynman diagrams, which may not follow the usual relationship of kinetic energy velocity in the end.
|
{
"evidence": [
"Virtual particle_sentence_3",
"Virtual particle_sentence_4",
"Virtual particle_sentence_19"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Virtual particle's concept arises in the perturbation theory of quantum field theory, where the calculations are often performed using schematic representations known as Feynman diagrams, which may not follow the usual relationship of kinetic energy velocity in the end.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Virtual particle_sentence_3",
"Virtual particle_sentence_4",
"Virtual particle_sentence_19"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6774,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.874",
"value": "?search="
},
{
"operation": "search",
"time": "27.325",
"value": "Virtual particle's concept"
},
{
"operation": "Now on",
"time": "33.353",
"value": "undefined"
},
{
"operation": "search",
"time": "45.45",
"value": "virtual particle"
},
{
"operation": "Now on",
"time": "46.128",
"value": "Virtual particle"
},
{
"operation": "Highlighting",
"time": "72.601",
"value": "Virtual particle_sentence_3"
},
{
"operation": "Highlighting",
"time": "73.924",
"value": "Virtual particle_sentence_4"
},
{
"operation": "hyperlink",
"time": "98.155",
"value": "Feynman diagram"
},
{
"operation": "Now on",
"time": "112.846",
"value": "Feynman diagram"
},
{
"operation": "Page search",
"time": "119.458",
"value": "kinetic velocity"
},
{
"operation": "back-button-clicked",
"time": "130.68",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "130.874",
"value": "Virtual particle"
},
{
"operation": "Page search",
"time": "138.246",
"value": "kinetic velocity"
},
{
"operation": "Highlighting",
"time": "149.142",
"value": "Virtual particle_sentence_19"
},
{
"operation": "finish",
"time": "160.058",
"value": "finish"
}
]
}
|
exfever_7220
|
Temple Grandin is the only award-winning film that premiered on HBO.
|
{
"evidence": [
"Temple Grandin (film)",
"Claire Danes",
"Hollywood Walk of Fame"
],
"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:\nTemple Grandin is the only award-winning film that premiered on HBO.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Temple Grandin (film)",
"Claire Danes",
"Hollywood Walk of Fame"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6775,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5991
|
Alkaline Trio is a British rock band from Chicago consisted of an American singer and musician and other band member.
|
{
"evidence": [
"Alkaline Trio",
"Dan Andriano"
],
"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:\nAlkaline Trio is a British rock band from Chicago consisted of an American singer and musician and other band member.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Alkaline Trio",
"Dan Andriano"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6776,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
43322
|
Speed skating in 2007 included the Fall World Cup Qualifier which featured five events for both men and women.
|
{
"evidence": [
"2006–07 ISU Speed Skating World Cup_sentence_0",
"2006–07 ISU Speed Skating World Cup_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:\nSpeed skating in 2007 included the Fall World Cup Qualifier which featured five events for both men and women.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"2006–07 ISU Speed Skating World Cup_sentence_0",
"2006–07 ISU Speed Skating World Cup_sentence_1"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6777,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.834",
"value": "?search="
},
{
"operation": "search",
"time": "14.366",
"value": "Speed skating in 2007"
},
{
"operation": "Now on",
"time": "16.042",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "21.228",
"value": "2007 World Cup"
},
{
"operation": "Now on",
"time": "22.456",
"value": "2007 World Cup"
},
{
"operation": "search",
"time": "39.789",
"value": "2007 Fall World Cup Qualifier"
},
{
"operation": "Now on",
"time": "41.228",
"value": "undefined"
},
{
"operation": "search",
"time": "73.884",
"value": "Speed skating at 2007 Fall World Cup"
},
{
"operation": "Now on",
"time": "75.041",
"value": "undefined"
},
{
"operation": "search",
"time": "95.593",
"value": "speed skating 2007"
},
{
"operation": "Now on",
"time": "96.525",
"value": "undefined"
},
{
"operation": "search",
"time": "113.245",
"value": "2007 Fall World Cup Qualifier"
},
{
"operation": "Now on",
"time": "113.78",
"value": "undefined"
},
{
"operation": "search",
"time": "132.349",
"value": "speed skating"
},
{
"operation": "Now on",
"time": "133.422",
"value": "Speed skating"
},
{
"operation": "Page search",
"time": "138.786",
"value": "2007"
},
{
"operation": "search",
"time": "170.09",
"value": "Fall World Cup 2007"
},
{
"operation": "Now on",
"time": "171.184",
"value": "undefined"
},
{
"operation": "search",
"time": "194.886",
"value": "SPeed skating at 2007 world cup"
},
{
"operation": "Now on",
"time": "195.861",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "199.616",
"value": "2007 World Cup"
},
{
"operation": "Now on",
"time": "200.017",
"value": "2007 World Cup"
},
{
"operation": "hyperlink",
"time": "204.45",
"value": "2007 Speed Skating World Cup"
},
{
"operation": "Now on",
"time": "205.011",
"value": "2006–07 ISU Speed Skating World Cup"
},
{
"operation": "hyperlink",
"time": "230.054",
"value": "ISU Speed Skating World Cup"
},
{
"operation": "Now on",
"time": "232.169",
"value": "ISU Speed Skating World Cup"
},
{
"operation": "hyperlink",
"time": "244.158",
"value": "2006–07"
},
{
"operation": "Now on",
"time": "244.94",
"value": "2006–07 ISU Speed Skating World Cup"
},
{
"operation": "Highlighting",
"time": "313.162",
"value": "2006–07 ISU Speed Skating World Cup_sentence_0"
},
{
"operation": "Highlighting",
"time": "314.609",
"value": "2006–07 ISU Speed Skating World Cup_sentence_1"
},
{
"operation": "finish",
"time": "321.568",
"value": "finish"
}
]
}
|
47826
|
Antony Valentini has been described as an "ardent admirer of de Broglie," which he emphasized that de Broglie, with the assistance of Erwin Schrödinger, a Nobel Prize-winning Austrian-Irish physicist, had constructed pilot-wave theory but later abandoned it in favor of quantum formalism.
|
{
"evidence": [
"Antony Valentini_sentence_11",
"Antony Valentini_sentence_13",
"Erwin Schrödinger_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:\nAntony Valentini has been described as an \"ardent admirer of de Broglie,\" which he emphasized that de Broglie, with the assistance of Erwin Schrödinger, a Nobel Prize-winning Austrian-Irish physicist, had constructed pilot-wave theory but later abandoned it in favor of quantum formalism.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Antony Valentini_sentence_11",
"Antony Valentini_sentence_13",
"Erwin Schrödinger_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6778,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.106",
"value": "?search="
},
{
"operation": "search",
"time": "11.678",
"value": "Antony Valentini"
},
{
"operation": "Now on",
"time": "13.232",
"value": "Antony Valentini"
},
{
"operation": "Page search",
"time": "30.874",
"value": "broglie"
},
{
"operation": "Highlighting",
"time": "46.974",
"value": "Antony Valentini_sentence_11"
},
{
"operation": "Highlighting",
"time": "50.485",
"value": "Antony Valentini_sentence_12"
},
{
"operation": "Highlighting",
"time": "55.868",
"value": "Antony Valentini_sentence_13"
},
{
"operation": "Highlighting deleted",
"time": "56.529",
"value": "Antony Valentini_sentence_12"
},
{
"operation": "hyperlink",
"time": "68.817",
"value": "Erwin Schrödinger"
},
{
"operation": "Now on",
"time": "70.721",
"value": "Erwin Schrödinger"
},
{
"operation": "Highlighting",
"time": "77.305",
"value": "Erwin Schrödinger_sentence_2"
},
{
"operation": "finish",
"time": "114.781",
"value": "finish"
}
]
}
|
exfever_5993
|
A calcaneal spur, also known as a prominence at the posterior end of the foot, is a bony outgrowth from the calcaneal tuberosity.
|
{
"evidence": [
"Calcaneal spur",
"Heel"
],
"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:\nA calcaneal spur, also known as a prominence at the posterior end of the foot, is a bony outgrowth from the calcaneal tuberosity.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Calcaneal spur",
"Heel"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6779,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
10459
|
Robert Lyminge was in charge of designing and construction of Hatfield House for Robert Cecil; then in 1616-17, he also designed the Blickling Hall for Sir Henry Hobart, 1st Baronet, an English politician who succeeded Sir Edward Coke to become Chief Justice of the Court of Common Pleas (1613-1625).
|
{
"evidence": [
"Robert Lyminge_sentence_3",
"Robert Lyminge_sentence_5",
"Sir Henry Hobart, 1st Baronet_sentence_1",
"Sir Henry Hobart, 1st Baronet_sentence_30"
],
"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:\nRobert Lyminge was in charge of designing and construction of Hatfield House for Robert Cecil; then in 1616-17, he also designed the Blickling Hall for Sir Henry Hobart, 1st Baronet, an English politician who succeeded Sir Edward Coke to become Chief Justice of the Court of Common Pleas (1613-1625).\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Robert Lyminge_sentence_3",
"Robert Lyminge_sentence_5",
"Sir Henry Hobart, 1st Baronet_sentence_1",
"Sir Henry Hobart, 1st Baronet_sentence_30"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6780,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.705",
"value": "?search="
},
{
"operation": "search",
"time": "23.531",
"value": "Robert Lyminge"
},
{
"operation": "Now on",
"time": "24.561",
"value": "Robert Lyminge"
},
{
"operation": "Highlighting",
"time": "66.276",
"value": "Robert Lyminge_sentence_3"
},
{
"operation": "Highlighting",
"time": "82.722",
"value": "Robert Lyminge_sentence_5"
},
{
"operation": "hyperlink",
"time": "122.716",
"value": "Sir Henry Hobart, 1st Baronet"
},
{
"operation": "Now on",
"time": "123.566",
"value": "Sir Henry Hobart, 1st Baronet"
},
{
"operation": "Highlighting",
"time": "139.314",
"value": "Sir Henry Hobart, 1st Baronet_sentence_1"
},
{
"operation": "back-button-clicked",
"time": "157.853",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "158.031",
"value": "Robert Lyminge"
},
{
"operation": "back-button-clicked",
"time": "176.02",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "176.03",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "176.208",
"value": "?search="
},
{
"operation": "search",
"time": "190.614",
"value": "Robert Lyminge"
},
{
"operation": "Now on",
"time": "191.634",
"value": "Robert Lyminge"
},
{
"operation": "back-button-clicked",
"time": "276.876",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "277.05",
"value": "?search="
},
{
"operation": "search",
"time": "287.692",
"value": "Robert Lyminge"
},
{
"operation": "Now on",
"time": "288.403",
"value": "Robert Lyminge"
},
{
"operation": "hyperlink",
"time": "367.706",
"value": "Sir Henry Hobart, 1st Baronet"
},
{
"operation": "Now on",
"time": "368.213",
"value": "Sir Henry Hobart, 1st Baronet"
},
{
"operation": "Highlighting",
"time": "399.018",
"value": "Sir Henry Hobart, 1st Baronet_sentence_30"
},
{
"operation": "finish",
"time": "721.012",
"value": "finish"
}
]
}
|
87820
|
Everett Glass was on the illustration staff of the Amherst Monthly and worked with Richard Ordynski in producing Henry IV.
|
{
"evidence": [
"Everett Glass_sentence_2"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nEverett Glass was on the illustration staff of the Amherst Monthly and worked with Richard Ordynski in producing Henry IV.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Everett Glass_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6781,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.853",
"value": "?search="
},
{
"operation": "search",
"time": "11.883",
"value": "Everett Glass"
},
{
"operation": "Now on",
"time": "12.846",
"value": "Everett Glass"
},
{
"operation": "Highlighting",
"time": "21.372",
"value": "Everett Glass_sentence_2"
},
{
"operation": "finish",
"time": "68.849",
"value": "finish"
}
]
}
|
90657
|
Arisan Maru was first assigned to transport 6,000 troops of the Kwantung Army from Pusan, Korea to Okinawa and was then ordered to transport POWs from the Philippines to Manchuria, China, or Japan.
|
{
"evidence": [
"Arisan Maru_sentence_12",
"Arisan Maru_sentence_13"
],
"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:\nArisan Maru was first assigned to transport 6,000 troops of the Kwantung Army from Pusan, Korea to Okinawa and was then ordered to transport POWs from the Philippines to Manchuria, China, or Japan.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Arisan Maru_sentence_12",
"Arisan Maru_sentence_13"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6782,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "4.101",
"value": "?search="
},
{
"operation": "search",
"time": "19.118",
"value": "Arisan Maru"
},
{
"operation": "Now on",
"time": "20.252",
"value": "Arisan Maru"
},
{
"operation": "Page search",
"time": "33.918",
"value": "kwantung"
},
{
"operation": "Highlighting",
"time": "40.531",
"value": "Arisan Maru_sentence_12"
},
{
"operation": "Highlighting",
"time": "51.893",
"value": "Arisan Maru_sentence_13"
},
{
"operation": "finish",
"time": "67.88",
"value": "finish"
}
]
}
|
25161
|
A species of sea snail is characterized by its long ribs and a trench which cuts through the ribs.
|
{
"evidence": [
"Gastropoda_sentence_2",
"Gastropoda_sentence_0",
"Invertebrate_sentence_0",
"Snail_sentence_35",
"Snail_sentence_36",
"Snail_sentence_3",
"Snail_sentence_1",
"Snail_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:\nA species of sea snail is characterized by its long ribs and a trench which cuts through the ribs.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Gastropoda_sentence_2",
"Gastropoda_sentence_0",
"Invertebrate_sentence_0",
"Snail_sentence_35",
"Snail_sentence_36",
"Snail_sentence_3",
"Snail_sentence_1",
"Snail_sentence_6"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6783,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.865",
"value": "?search="
},
{
"operation": "search",
"time": "10.392",
"value": "sea snail"
},
{
"operation": "Now on",
"time": "11.698",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "15.719",
"value": "Types of snails by habitat Snail_section_2"
},
{
"operation": "Now on",
"time": "16.962",
"value": "Snail#Types of snails by habitat Snail section 2"
},
{
"operation": "Highlighting",
"time": "25.419",
"value": "Snail_sentence_35"
},
{
"operation": "hyperlink",
"time": "35.35",
"value": "Snail (disambiguation)"
},
{
"operation": "Now on",
"time": "36.337",
"value": "Snail (disambiguation)"
},
{
"operation": "hyperlink",
"time": "51.449",
"value": "Gastropoda"
},
{
"operation": "Now on",
"time": "52.766",
"value": "Gastropoda"
},
{
"operation": "Highlighting",
"time": "61.158",
"value": "Gastropoda_sentence_2"
},
{
"operation": "Page search",
"time": "87.066",
"value": "sea"
},
{
"operation": "Highlighting",
"time": "112.375",
"value": "Gastropoda_sentence_13"
},
{
"operation": "Page search",
"time": "139.427",
"value": "trench"
},
{
"operation": "Highlighting deleted",
"time": "142.156",
"value": "Gastropoda_sentence_2"
},
{
"operation": "Highlighting",
"time": "143.092",
"value": "Gastropoda_sentence_2"
},
{
"operation": "Page search",
"time": "167.597",
"value": "rib"
},
{
"operation": "Highlighting",
"time": "194.084",
"value": "Gastropoda_sentence_0"
},
{
"operation": "hyperlink",
"time": "195.163",
"value": "invertebrates"
},
{
"operation": "Now on",
"time": "196.638",
"value": "Invertebrate"
},
{
"operation": "Highlighting",
"time": "203.259",
"value": "Invertebrate_sentence_0"
},
{
"operation": "search",
"time": "254.114",
"value": "sea snail"
},
{
"operation": "Now on",
"time": "255.174",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "257.136",
"value": "Types of snails by habitat Snail_section_2"
},
{
"operation": "Now on",
"time": "257.986",
"value": "Snail#Types of snails by habitat Snail section 2"
},
{
"operation": "Highlighting",
"time": "259.947",
"value": "Snail_sentence_35"
},
{
"operation": "Highlighting",
"time": "262.007",
"value": "Snail_sentence_36"
},
{
"operation": "Highlighting",
"time": "271.041",
"value": "Snail_sentence_3"
},
{
"operation": "Highlighting",
"time": "277.462",
"value": "Snail_sentence_1"
},
{
"operation": "Highlighting",
"time": "285.43",
"value": "Snail_sentence_6"
},
{
"operation": "hyperlink",
"time": "298.824",
"value": "3 Types of snails by habitat Snail_section_2"
},
{
"operation": "Now on",
"time": "318.746",
"value": "Snail#Types of snails by habitat Snail section 2"
},
{
"operation": "finish",
"time": "335.562",
"value": "finish"
}
]
}
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.