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 |
|---|---|---|---|---|---|---|---|---|
52036
|
The Daily Observer ran the history corner pionnered by Alieu Ebrima Cham Joof on the weekend was eventually shutdown by tax authorities
|
{
"evidence": [
"Alieu Ebrima Cham Joof_sentence_140",
"The Daily Observer_sentence_5"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Daily Observer ran the history corner pionnered by Alieu Ebrima Cham Joof on the weekend was eventually shutdown by tax authorities\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Alieu Ebrima Cham Joof_sentence_140",
"The Daily Observer_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2984,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.134",
"value": "?search="
},
{
"operation": "search",
"time": "21.908",
"value": "Daily Observer"
},
{
"operation": "Now on",
"time": "22.707",
"value": "Daily Observer"
},
{
"operation": "search",
"time": "43.659",
"value": "Alieu Ebrima Cham Joof"
},
{
"operation": "Now on",
"time": "44.837",
"value": "Alieu Ebrima Cham Joof"
},
{
"operation": "Page search",
"time": "50.766",
"value": "observer"
},
{
"operation": "Highlighting",
"time": "85.715",
"value": "Alieu Ebrima Cham Joof_sentence_140"
},
{
"operation": "hyperlink",
"time": "105.578",
"value": "The Daily Observer"
},
{
"operation": "Now on",
"time": "106.349",
"value": "The Daily Observer"
},
{
"operation": "Page search",
"time": "122.043",
"value": "tax"
},
{
"operation": "Highlighting",
"time": "124.666",
"value": "The Daily Observer_sentence_5"
},
{
"operation": "finish",
"time": "347.726",
"value": "finish"
}
]
}
|
exfever_1178
|
French is a Romance language of the Indo-European family that evolved from Gallo-Romance, the Latin spoke in a region in Western Europe that was inhabited by Celtic tribes, encompassing present-day France.
|
{
"evidence": [
"French language",
"Gaul"
],
"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:\nFrench is a Romance language of the Indo-European family that evolved from Gallo-Romance, the Latin spoke in a region in Western Europe that was inhabited by Celtic tribes, encompassing present-day France.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"French language",
"Gaul"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2985,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3534
|
Amy Poehler is an American actress who studied with an improvisational comedy enterprise and is the first ongoing improvisational theater troupe to be continually based in Chicago.
|
{
"evidence": [
"Amy Poehler",
"The Second City"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nAmy Poehler is an American actress who studied with an improvisational comedy enterprise and is the first ongoing improvisational theater troupe to be continually based in Chicago.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Amy Poehler",
"The Second City"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2986,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7709
|
Notes on a Scandal is a 2006 British psychological thriller-drama film adapted from the 2003 short story by Zoë Heller.
|
{
"evidence": [
"Notes on a Scandal (film)",
"Notes on a Scandal"
],
"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:\nNotes on a Scandal is a 2006 British psychological thriller-drama film adapted from the 2003 short story by Zoë Heller.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Notes on a Scandal (film)",
"Notes on a Scandal"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2987,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
26031
|
Radochów, which was located in Germany before being recognized as being in south-western Poland, has a population of 500.
|
{
"evidence": [
"Radochów_sentence_1",
"Radochów_sentence_0",
"Radochów_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:\nRadochów, which was located in Germany before being recognized as being in south-western Poland, has a population of 500.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Radochów_sentence_1",
"Radochów_sentence_0",
"Radochów_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2988,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.582",
"value": "?search="
},
{
"operation": "search",
"time": "5.143",
"value": "Radochow"
},
{
"operation": "Now on",
"time": "5.839",
"value": "Radochów"
},
{
"operation": "Highlighting",
"time": "19.265",
"value": "Radochów_sentence_1"
},
{
"operation": "Highlighting",
"time": "25.069",
"value": "Radochów_sentence_0"
},
{
"operation": "Highlighting",
"time": "28.512",
"value": "Radochów_sentence_3"
},
{
"operation": "finish",
"time": "44.172",
"value": "finish"
}
]
}
|
exfever_2447
|
Rihanna gained recognition with the release of her albums, Girl Like Me and her debut studio album, which were released by an Australian multinational record label owned by the Universal Music Group.
|
{
"evidence": [
"Rihanna",
"Music of the Sun",
"Def Jam Recordings"
],
"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:\nRihanna gained recognition with the release of her albums, Girl Like Me and her debut studio album, which were released by an Australian multinational record label owned by the Universal Music Group.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Rihanna",
"Music of the Sun",
"Def Jam Recordings"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2989,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
95539
|
Christopher Musonda, born in 1986, in Zambia (which is in Southern-Central Africa), has never played football.
|
{
"evidence": [
"Christopher Musonda_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:\nChristopher Musonda, born in 1986, in Zambia (which is in Southern-Central Africa), has never played football.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Christopher Musonda_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2990,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.948",
"value": "?search="
},
{
"operation": "search",
"time": "9.656",
"value": "Christopher Musonda"
},
{
"operation": "Now on",
"time": "10.462",
"value": "Christopher Musonda"
},
{
"operation": "Highlighting",
"time": "26.166",
"value": "Christopher Musonda_header_cell_0_7_0"
},
{
"operation": "Highlighting",
"time": "27.693",
"value": "Christopher Musonda_cell_0_7_1"
},
{
"operation": "Highlighting",
"time": "27.711",
"value": "Christopher Musonda_header_cell_0_7_0"
},
{
"operation": "Highlighting",
"time": "27.712",
"value": "Christopher Musonda_header_cell_0_5_0"
},
{
"operation": "hyperlink",
"time": "29.03",
"value": "Forest Rangers"
},
{
"operation": "Now on",
"time": "29.956",
"value": "Forest Rangers F.C."
},
{
"operation": "Highlighting",
"time": "34.35",
"value": "Forest Rangers F.C._sentence_0"
},
{
"operation": "Highlighting",
"time": "34.352",
"value": "Forest Rangers F.C._title"
},
{
"operation": "back-button-clicked",
"time": "43.958",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "44.222",
"value": "Christopher Musonda"
},
{
"operation": "Highlighting deleted",
"time": "56.776",
"value": "Christopher Musonda_cell_0_7_1"
},
{
"operation": "Highlighting deleted",
"time": "56.787",
"value": "Christopher Musonda_header_cell_0_7_0"
},
{
"operation": "Highlighting deleted",
"time": "56.788",
"value": "Christopher Musonda_header_cell_0_5_0"
},
{
"operation": "Highlighting",
"time": "58.407",
"value": "Christopher Musonda_sentence_0"
},
{
"operation": "finish",
"time": "71.07",
"value": "finish"
}
]
}
|
exfever_3613
|
Measles is a highly contagious infectious disease with complications including an inflammatory condition of the lung primarily affecting the small air sacs known as alveoli.
|
{
"evidence": [
"Measles",
"Pneumonia"
],
"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:\nMeasles is a highly contagious infectious disease with complications including an inflammatory condition of the lung primarily affecting the small air sacs known as alveoli.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Measles",
"Pneumonia"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2991,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
12366
|
Ramsden, a British painter who lives but never works in London, studied art at Leith School of Art (LSA) and Edinburgh College of Art, then continued her education and graduated in 2013 from the three-year postgraduate art school at the Royal Academy of Arts, RA, in London.
|
{
"evidence": [
"Mary Ramsden_sentence_0",
"Mary Ramsden_sentence_2",
"Mary Ramsden_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:\nRamsden, a British painter who lives but never works in London, studied art at Leith School of Art (LSA) and Edinburgh College of Art, then continued her education and graduated in 2013 from the three-year postgraduate art school at the Royal Academy of Arts, RA, in London.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Mary Ramsden_sentence_0",
"Mary Ramsden_sentence_2",
"Mary Ramsden_sentence_3"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2992,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "3.166",
"value": "?search="
},
{
"operation": "search",
"time": "16.114",
"value": "Ramsden"
},
{
"operation": "Now on",
"time": "18.239",
"value": "Ramsden"
},
{
"operation": "search",
"time": "36.338",
"value": "Ramsden painter"
},
{
"operation": "Now on",
"time": "38.33",
"value": "undefined"
},
{
"operation": "search",
"time": "65.735",
"value": "ramsden artist"
},
{
"operation": "Now on",
"time": "66.568",
"value": "undefined"
},
{
"operation": "search",
"time": "107.022",
"value": "mary ramsden"
},
{
"operation": "Now on",
"time": "108.287",
"value": "Mary Ramsden"
},
{
"operation": "Highlighting",
"time": "118.928",
"value": "Mary Ramsden_sentence_0"
},
{
"operation": "Highlighting",
"time": "134.321",
"value": "Mary Ramsden_sentence_2"
},
{
"operation": "Highlighting",
"time": "147.929",
"value": "Mary Ramsden_sentence_3"
},
{
"operation": "finish",
"time": "169.864",
"value": "finish"
}
]
}
|
exfever_2920
|
Smoking is the no.1 cause of Leukemia.
|
{
"evidence": [
"Leukemia",
"Bone pain"
],
"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:\nSmoking is the no.1 cause of Leukemia.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Leukemia",
"Bone pain"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2993,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
2111
|
Lincheng County, which is located in the province Hebei in China, in the foothills of the Taihang Mountains, has a total area of 308 square miles and a population of 204,000.
|
{
"evidence": [
"Lincheng County_sentence_0",
"Lincheng County_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:\nLincheng County, which is located in the province Hebei in China, in the foothills of the Taihang Mountains, has a total area of 308 square miles and a population of 204,000.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Lincheng County_sentence_0",
"Lincheng County_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2994,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "search",
"time": "26.773",
"value": "Lincheng County"
},
{
"operation": "Highlighting",
"time": "34.817",
"value": "Lincheng County_sentence_0"
},
{
"operation": "Highlighting",
"time": "40.876",
"value": "Lincheng County_sentence_2"
},
{
"operation": "finish",
"time": "69.231",
"value": "finish"
}
]
}
|
53801
|
Operation Dragnet (May 26, 1967 – January 27, 1968)was a security operation that used the tactic later known as Unicorn Program and was conducted by the 1st Cavalry Division in Bình Định Province.
|
{
"evidence": [
"Operation Dragnet_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:\nOperation Dragnet (May 26, 1967 – January 27, 1968)was a security operation that used the tactic later known as Unicorn Program and was conducted by the 1st Cavalry Division in Bình Định Province.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Operation Dragnet_sentence_6"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2995,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.48",
"value": "?search="
},
{
"operation": "search",
"time": "12.826",
"value": "Operation Dragnet"
},
{
"operation": "Now on",
"time": "15.08",
"value": "Operation Dragnet"
},
{
"operation": "Highlighting",
"time": "325.51",
"value": "Operation Dragnet_sentence_0"
},
{
"operation": "Page search",
"time": "349.013",
"value": "unicorn"
},
{
"operation": "Highlighting",
"time": "357.49",
"value": "Operation Dragnet_sentence_6"
},
{
"operation": "Highlighting deleted",
"time": "365.552",
"value": "Operation Dragnet_sentence_0"
},
{
"operation": "finish",
"time": "372.92",
"value": "finish"
}
]
}
|
72641
|
Vuzmetinci, Slovenia, located in Northern Europe, has a total population of 140.
|
{
"evidence": [
"Slovenia_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:\nVuzmetinci, Slovenia, located in Northern Europe, has a total population of 140.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Slovenia_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2996,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.821",
"value": "?search="
},
{
"operation": "search",
"time": "25.148",
"value": "Vuzmetinci"
},
{
"operation": "Now on",
"time": "25.943",
"value": "Vuzmetinci"
},
{
"operation": "Highlighting",
"time": "30.039",
"value": "Vuzmetinci_cell_0_2_1"
},
{
"operation": "Highlighting",
"time": "30.039",
"value": "Vuzmetinci_title"
},
{
"operation": "Highlighting",
"time": "30.045",
"value": "Vuzmetinci_header_cell_0_2_0"
},
{
"operation": "Highlighting",
"time": "30.045",
"value": "Vuzmetinci_header_cell_0_0_0"
},
{
"operation": "Highlighting deleted",
"time": "30.731",
"value": "Vuzmetinci_cell_0_2_1"
},
{
"operation": "Highlighting deleted",
"time": "30.732",
"value": "Vuzmetinci_title"
},
{
"operation": "Highlighting deleted",
"time": "30.74",
"value": "Vuzmetinci_header_cell_0_2_0"
},
{
"operation": "Highlighting deleted",
"time": "30.741",
"value": "Vuzmetinci_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "33.274",
"value": "Vuzmetinci_cell_0_1_1"
},
{
"operation": "Highlighting",
"time": "33.275",
"value": "Vuzmetinci_title"
},
{
"operation": "Highlighting",
"time": "33.279",
"value": "Vuzmetinci_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "33.279",
"value": "Vuzmetinci_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "36.625",
"value": "Vuzmetinci_cell_0_9_1"
},
{
"operation": "Highlighting",
"time": "36.63",
"value": "Vuzmetinci_header_cell_0_9_0"
},
{
"operation": "Highlighting",
"time": "36.631",
"value": "Vuzmetinci_header_cell_0_8_0"
},
{
"operation": "hyperlink",
"time": "55.223",
"value": "Slovenia"
},
{
"operation": "Now on",
"time": "70.564",
"value": "Vuzmetinci"
},
{
"operation": "hyperlink",
"time": "72.499",
"value": "Slovenia"
},
{
"operation": "Now on",
"time": "75.614",
"value": "Slovenia"
},
{
"operation": "Highlighting",
"time": "78.624",
"value": "Slovenia_sentence_1"
},
{
"operation": "finish",
"time": "92.742",
"value": "finish"
}
]
}
|
exfever_7922
|
That's So Raven is Anneliese van der Pol's second television series.
|
{
"evidence": [
"Anneliese van der Pol",
"5 Weddings"
],
"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:\nThat's So Raven is Anneliese van der Pol's second television series.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Anneliese van der Pol",
"5 Weddings"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2997,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
64241
|
All the published writings of Louisa Parr was issued under the pen name of Mrs Olinthus Lobb.
|
{
"evidence": [
"Louisa Parr_sentence_7"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nAll the published writings of Louisa Parr was issued under the pen name of Mrs Olinthus Lobb.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Louisa Parr_sentence_7"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2998,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.119",
"value": "?search="
},
{
"operation": "Page search",
"time": "17.991",
"value": "Louisa Parr"
},
{
"operation": "search",
"time": "20.91",
"value": "Louisa Parr"
},
{
"operation": "Now on",
"time": "22.064",
"value": "Louisa Parr"
},
{
"operation": "Highlighting",
"time": "35.373",
"value": "Louisa Parr_sentence_1"
},
{
"operation": "Highlighting",
"time": "37.512",
"value": "Louisa Parr_sentence_0"
},
{
"operation": "Highlighting",
"time": "48.479",
"value": "Louisa Parr_sentence_7"
},
{
"operation": "Highlighting deleted",
"time": "51.149",
"value": "Louisa Parr_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "52.411",
"value": "Louisa Parr_sentence_0"
},
{
"operation": "finish",
"time": "60.473",
"value": "finish"
}
]
}
|
exfever_7493
|
German is the most widely spoken mother language and an official language in four countries in the European Union including Germany, Austria, Belgium, and Luxembourg.
|
{
"evidence": [
"Luxembourg",
"French language"
],
"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:\nGerman is the most widely spoken mother language and an official language in four countries in the European Union including Germany, Austria, Belgium, and Luxembourg.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Luxembourg",
"French language"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2999,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
92619
|
John Druce(born February 23, 1966) in Australia is a former right wing ice hockey player who played for Washington Capitals, Winnipeg Jets, Los Angeles Kings and Philadelphia Flyers in the NHL and is now currently working for Freedom 55 Financial in Peterborough.
|
{
"evidence": [
"John Druce_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:\nJohn Druce(born February 23, 1966) in Australia is a former right wing ice hockey player who played for Washington Capitals, Winnipeg Jets, Los Angeles Kings and Philadelphia Flyers in the NHL and is now currently working for Freedom 55 Financial in Peterborough.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"John Druce_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3000,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.441",
"value": "?search="
},
{
"operation": "search",
"time": "6.39",
"value": "John Druce"
},
{
"operation": "Now on",
"time": "7.223",
"value": "John Druce"
},
{
"operation": "Highlighting",
"time": "11.965",
"value": "John Druce_sentence_0"
},
{
"operation": "finish",
"time": "24.888",
"value": "finish"
}
]
}
|
52323
|
Venancio Ramos joined French side Lens before returning to South America in 1986 to play for Independiente of Argentina and he returned to Uruguay in 1989 where he played for Racing Montevideo then Peñarol's fiercest rivals Nacional.
|
{
"evidence": [
"Venancio Ramos_sentence_2",
"Venancio Ramos_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:\nVenancio Ramos joined French side Lens before returning to South America in 1986 to play for Independiente of Argentina and he returned to Uruguay in 1989 where he played for Racing Montevideo then Peñarol's fiercest rivals Nacional.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Venancio Ramos_sentence_2",
"Venancio Ramos_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3001,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.455",
"value": "?search="
},
{
"operation": "search",
"time": "8.929",
"value": "Venancio Ramos"
},
{
"operation": "Now on",
"time": "10.081",
"value": "Venancio Ramos"
},
{
"operation": "Highlighting",
"time": "35.828",
"value": "Venancio Ramos_sentence_2"
},
{
"operation": "Highlighting",
"time": "39.867",
"value": "Venancio Ramos_sentence_3"
},
{
"operation": "finish",
"time": "48.197",
"value": "finish"
}
]
}
|
38051
|
The Kocatepe Mosque, the largest mosque in Ankara, Turkey, was built with the help of Ahmet Hamdi Akseki and the founding members of the "Society to Build a Mosque in Yenişehir, Ankara.”
|
{
"evidence": [
"Kocatepe Mosque_sentence_0",
"Kocatepe Mosque_sentence_1",
"Kocatepe Mosque_sentence_3",
"Kocatepe Mosque_sentence_4",
"Kocatepe Mosque_sentence_5",
"Kocatepe Mosque_sentence_6"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Kocatepe Mosque, the largest mosque in Ankara, Turkey, was built with the help of Ahmet Hamdi Akseki and the founding members of the \"Society to Build a Mosque in Yenişehir, Ankara.”\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Kocatepe Mosque_sentence_0",
"Kocatepe Mosque_sentence_1",
"Kocatepe Mosque_sentence_3",
"Kocatepe Mosque_sentence_4",
"Kocatepe Mosque_sentence_5",
"Kocatepe Mosque_sentence_6"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3002,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.166",
"value": "undefined"
},
{
"operation": "search",
"time": "18.288",
"value": "Kocatepe Mosque"
},
{
"operation": "Now on",
"time": "19.833",
"value": "Kocatepe Mosque"
},
{
"operation": "Highlighting",
"time": "25.648",
"value": "Kocatepe Mosque_sentence_0"
},
{
"operation": "Highlighting",
"time": "36.858",
"value": "Kocatepe Mosque_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "41.423",
"value": "Kocatepe Mosque_sentence_1"
},
{
"operation": "Page search",
"time": "109.105",
"value": "ahmet"
},
{
"operation": "Page search",
"time": "109.263",
"value": "ahmet"
},
{
"operation": "Page search",
"time": "109.432",
"value": "ahmet"
},
{
"operation": "Page search",
"time": "109.594",
"value": "ahmet"
},
{
"operation": "Page search",
"time": "109.783",
"value": "ahmet"
},
{
"operation": "Page search",
"time": "109.952",
"value": "ahmet"
},
{
"operation": "Highlighting",
"time": "123.721",
"value": "Kocatepe Mosque_sentence_3"
},
{
"operation": "Highlighting",
"time": "209.211",
"value": "Kocatepe Mosque_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "294.585",
"value": "Kocatepe Mosque_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "326.619",
"value": "Kocatepe Mosque_sentence_4"
},
{
"operation": "Highlighting",
"time": "328.048",
"value": "Kocatepe Mosque_sentence_0"
},
{
"operation": "Highlighting",
"time": "328.54",
"value": "Kocatepe Mosque_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "330.584",
"value": "Kocatepe Mosque_sentence_3"
},
{
"operation": "Highlighting",
"time": "335.349",
"value": "Kocatepe Mosque_sentence_3"
},
{
"operation": "Highlighting",
"time": "341.282",
"value": "Kocatepe Mosque_sentence_4"
},
{
"operation": "Highlighting",
"time": "363.429",
"value": "Kocatepe Mosque_sentence_5"
},
{
"operation": "Highlighting",
"time": "363.902",
"value": "Kocatepe Mosque_sentence_6"
},
{
"operation": "finish",
"time": "374.575",
"value": "finish"
}
]
}
|
93533
|
Simon Taylor, a barque built neither in 1824 nor in 1842 on the River Thames. brought 18 Parkhurst apprentices to Western Australia on August 1842; then after three months, St. George brought 92 Parkhurst apprentices to New Zealand.
|
{
"evidence": [
"Simon Taylor (ship)_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:\nSimon Taylor, a barque built neither in 1824 nor in 1842 on the River Thames. brought 18 Parkhurst apprentices to Western Australia on August 1842; then after three months, St. George brought 92 Parkhurst apprentices to New Zealand.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Simon Taylor (ship)_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3003,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.181",
"value": "?search="
},
{
"operation": "search",
"time": "20.122",
"value": "Simon Taylor"
},
{
"operation": "Now on",
"time": "20.785",
"value": "Simon Taylor"
},
{
"operation": "search",
"time": "100.002",
"value": "simon taylor barque"
},
{
"operation": "Now on",
"time": "101.82",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "105.493",
"value": "Simon Taylor (ship)"
},
{
"operation": "Now on",
"time": "105.816",
"value": "Simon Taylor (ship)"
},
{
"operation": "Highlighting",
"time": "107.27",
"value": "Simon Taylor (ship)_sentence_1"
},
{
"operation": "finish",
"time": "114.884",
"value": "finish"
}
]
}
|
26280
|
Byron Bay's own Skegss album Holiday Food was reviewed by three critics.
|
{
"evidence": [
"Skegss_sentence_0",
"Skegss_sentence_6",
"Skegss_sentence_27",
"Skegss_sentence_28",
"Skegss_sentence_37"
],
"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:\nByron Bay's own Skegss album Holiday Food was reviewed by three critics.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Skegss_sentence_0",
"Skegss_sentence_6",
"Skegss_sentence_27",
"Skegss_sentence_28",
"Skegss_sentence_37"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3004,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.358",
"value": "?search="
},
{
"operation": "search",
"time": "20.67",
"value": "Byron Bay Record"
},
{
"operation": "Now on",
"time": "21.681",
"value": "Byron Bay Record"
},
{
"operation": "search",
"time": "34.464",
"value": "Contains...Holiday Food"
},
{
"operation": "Now on",
"time": "36.128",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "38.428",
"value": "Skegss"
},
{
"operation": "Now on",
"time": "39.161",
"value": "Skegss"
},
{
"operation": "Highlighting",
"time": "45.115",
"value": "Skegss_sentence_0"
},
{
"operation": "Page search",
"time": "57.452",
"value": "Holiday Food"
},
{
"operation": "hyperlink",
"time": "71.963",
"value": "4 Discography Skegss_section_3"
},
{
"operation": "Highlighting",
"time": "114.815",
"value": "Skegss_sentence_31"
},
{
"operation": "Highlighting deleted",
"time": "120.523",
"value": "Skegss_sentence_31"
},
{
"operation": "search",
"time": "146.023",
"value": "Contains...\"holiday food\""
},
{
"operation": "Now on",
"time": "147.641",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "161.498",
"value": "Skegss"
},
{
"operation": "Now on",
"time": "162.559",
"value": "Skegss"
},
{
"operation": "Page search",
"time": "180.503",
"value": "holiday food"
},
{
"operation": "Highlighting",
"time": "184.966",
"value": "Skegss_sentence_6"
},
{
"operation": "Highlighting",
"time": "194.122",
"value": "Skegss_sentence_27"
},
{
"operation": "Highlighting",
"time": "198.385",
"value": "Skegss_sentence_28"
},
{
"operation": "Highlighting",
"time": "287.775",
"value": "Skegss_sentence_37"
},
{
"operation": "Now on",
"time": "332.17",
"value": "Skegss"
},
{
"operation": "finish",
"time": "362.032",
"value": "finish"
}
]
}
|
22000
|
Out of the four competitors in the Dominican delegation at the 2011 Commonwealth Youth Games two were women , Chelsey Linton who ranked first place for her high jump, and Shanica Yankey who also won a medal for her javelin throw.
|
{
"evidence": [
"Dominica at the 2011 Commonwealth Youth Games_sentence_15"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nOut of the four competitors in the Dominican delegation at the 2011 Commonwealth Youth Games two were women , Chelsey Linton who ranked first place for her high jump, and Shanica Yankey who also won a medal for her javelin throw.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Dominica at the 2011 Commonwealth Youth Games_sentence_15"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3005,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.96",
"value": "?search="
},
{
"operation": "search",
"time": "14.23",
"value": "Contains...2011 Commenwealth Youth Games"
},
{
"operation": "Now on",
"time": "15.815",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "17.77",
"value": "2011 Commonwealth Youth Games"
},
{
"operation": "Now on",
"time": "18.751",
"value": "2011 Commonwealth Youth Games"
},
{
"operation": "Page search",
"time": "34.769",
"value": "dominican"
},
{
"operation": "Page search",
"time": "52.009",
"value": "chelsey linton"
},
{
"operation": "Page search",
"time": "55.663",
"value": "linton"
},
{
"operation": "search",
"time": "71.511",
"value": "Contains...Chelsey Linton"
},
{
"operation": "Now on",
"time": "72.709",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "78.519",
"value": "Dominica at the 2011 Commonwealth Youth Games"
},
{
"operation": "Now on",
"time": "79.478",
"value": "Dominica at the 2011 Commonwealth Youth Games"
},
{
"operation": "Page search",
"time": "83.263",
"value": "chelsey"
},
{
"operation": "Highlighting",
"time": "100.464",
"value": "Dominica at the 2011 Commonwealth Youth Games_sentence_11"
},
{
"operation": "Highlighting",
"time": "100.465",
"value": "Dominica at the 2011 Commonwealth Youth Games_section_0"
},
{
"operation": "hyperlink",
"time": "138.083",
"value": "Athletics at the 2011 Commonwealth Youth Games"
},
{
"operation": "Now on",
"time": "139.617",
"value": "Athletics at the 2011 Commonwealth Youth Games"
},
{
"operation": "search",
"time": "158.676",
"value": "Dominica at the 2011 Commonwealth Youth Games"
},
{
"operation": "Now on",
"time": "159.184",
"value": "Dominica at the 2011 Commonwealth Youth Games"
},
{
"operation": "Highlighting",
"time": "221.27",
"value": "Dominica at the 2011 Commonwealth Youth Games_sentence_15"
},
{
"operation": "Highlighting deleted",
"time": "225.647",
"value": "Dominica at the 2011 Commonwealth Youth Games_sentence_15"
},
{
"operation": "Highlighting",
"time": "226.745",
"value": "Dominica at the 2011 Commonwealth Youth Games_sentence_15"
},
{
"operation": "Highlighting deleted",
"time": "232.58",
"value": "Dominica at the 2011 Commonwealth Youth Games_sentence_11"
},
{
"operation": "Highlighting deleted",
"time": "232.582",
"value": "Dominica at the 2011 Commonwealth Youth Games_section_0"
},
{
"operation": "finish",
"time": "262.979",
"value": "finish"
}
]
}
|
48240
|
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.
|
{
"evidence": [
"I Like How It Feels_sentence_0",
"Enrique Iglesias_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:\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.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"I Like How It Feels_sentence_0",
"Enrique Iglesias_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3006,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.988",
"value": "?search="
},
{
"operation": "search",
"time": "12.11",
"value": "I Like How It Feels"
},
{
"operation": "Now on",
"time": "13.319",
"value": "I Like How It Feels"
},
{
"operation": "Highlighting",
"time": "18.973",
"value": "I Like How It Feels_sentence_0"
},
{
"operation": "hyperlink",
"time": "24.362",
"value": "Enrique Iglesias"
},
{
"operation": "Now on",
"time": "25.719",
"value": "Enrique Iglesias"
},
{
"operation": "Highlighting",
"time": "28.33",
"value": "Enrique Iglesias_sentence_3"
},
{
"operation": "finish",
"time": "48.49",
"value": "finish"
}
]
}
|
3795
|
Tilian Pearson has been a vocalist for the bands Tides of Man and Dance Gavin Dance, who he toured with in February and March of 2017 along with the band Coheed and Cambria.
|
{
"evidence": [
"Tilian Pearson_sentence_1",
"Tilian Pearson_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:\nTilian Pearson has been a vocalist for the bands Tides of Man and Dance Gavin Dance, who he toured with in February and March of 2017 along with the band Coheed and Cambria.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Tilian Pearson_sentence_1",
"Tilian Pearson_sentence_2"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3007,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "search",
"time": "16.056",
"value": "Tilian Pearson"
},
{
"operation": "Highlighting",
"time": "40.453",
"value": "Tilian Pearson_sentence_1"
},
{
"operation": "Highlighting",
"time": "43.652",
"value": "Tilian Pearson_sentence_2"
},
{
"operation": "hyperlink",
"time": "63.46",
"value": "2.2 2017–present"
},
{
"operation": "hyperlink",
"time": "128.716",
"value": "Dance Gavin Dance"
},
{
"operation": "back-button-clicked",
"time": "224.822",
"value": "back-button-clicked"
},
{
"operation": "search",
"time": "252.693",
"value": "coheed"
},
{
"operation": "hyperlink",
"time": "278.996",
"value": "1.7 The Color Before the Sun (2015–2017) Coheed and Cambria_section_7"
},
{
"operation": "Page search",
"time": "304.548",
"value": "dance gavin dance"
},
{
"operation": "finish",
"time": "334.011",
"value": "finish"
}
]
}
|
exfever_8500
|
Barbara Pierce Bush was the first lady of the United States from 1989 to 1993 and the mother of an American politician that served as the 46th governor of Texas as part of the Grand Old Party, one of the two major contemporary political parties in the United States.
|
{
"evidence": [
"Barbara Bush",
"George W. Bush",
"Republican Party (United States)"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBarbara Pierce Bush was the first lady of the United States from 1989 to 1993 and the mother of an American politician that served as the 46th governor of Texas as part of the Grand Old Party, one of the two major contemporary political parties in the United States.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Barbara Bush",
"George W. Bush",
"Republican Party (United States)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3008,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
71281
|
The Landscape Agency was founded by James in October 1998 after teaming up with two associates and James grew the business by seeking additional start-up capital after a year.
|
{
"evidence": [
"The Landscape Agency_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 Landscape Agency was founded by James in October 1998 after teaming up with two associates and James grew the business by seeking additional start-up capital after a year.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"The Landscape Agency_sentence_6"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3009,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.74",
"value": "?search="
},
{
"operation": "search",
"time": "17.574",
"value": "Landscape Agency"
},
{
"operation": "Now on",
"time": "20.707",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "23.208",
"value": "The Landscape Agency"
},
{
"operation": "Now on",
"time": "23.791",
"value": "The Landscape Agency"
},
{
"operation": "Highlighting",
"time": "26.961",
"value": "The Landscape Agency_sentence_0"
},
{
"operation": "Highlighting",
"time": "32.395",
"value": "The Landscape Agency_table_caption_0"
},
{
"operation": "Highlighting",
"time": "36.891",
"value": "The Landscape Agency_cell_0_2_1"
},
{
"operation": "Highlighting",
"time": "36.894",
"value": "The Landscape Agency_header_cell_0_2_0"
},
{
"operation": "Highlighting",
"time": "37.527",
"value": "The Landscape Agency_cell_0_1_1"
},
{
"operation": "Highlighting",
"time": "37.53",
"value": "The Landscape Agency_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "58.932",
"value": "The Landscape Agency_sentence_7"
},
{
"operation": "Highlighting deleted",
"time": "59.46",
"value": "The Landscape Agency_sentence_7"
},
{
"operation": "Highlighting",
"time": "60.576",
"value": "The Landscape Agency_sentence_5"
},
{
"operation": "Highlighting deleted",
"time": "64.194",
"value": "The Landscape Agency_cell_0_2_1"
},
{
"operation": "Highlighting deleted",
"time": "64.198",
"value": "The Landscape Agency_header_cell_0_2_0"
},
{
"operation": "Highlighting deleted",
"time": "65.003",
"value": "The Landscape Agency_cell_0_1_1"
},
{
"operation": "Highlighting deleted",
"time": "65.006",
"value": "The Landscape Agency_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "65.747",
"value": "The Landscape Agency_cell_0_1_1"
},
{
"operation": "Highlighting",
"time": "65.749",
"value": "The Landscape Agency_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "66.838",
"value": "The Landscape Agency_cell_0_2_1"
},
{
"operation": "Highlighting",
"time": "66.841",
"value": "The Landscape Agency_header_cell_0_2_0"
},
{
"operation": "Highlighting deleted",
"time": "67.779",
"value": "The Landscape Agency_sentence_5"
},
{
"operation": "Highlighting",
"time": "68.168",
"value": "The Landscape Agency_sentence_5"
},
{
"operation": "Highlighting",
"time": "76.76",
"value": "The Landscape Agency_sentence_7"
},
{
"operation": "Highlighting deleted",
"time": "77.257",
"value": "The Landscape Agency_sentence_7"
},
{
"operation": "Highlighting",
"time": "79.239",
"value": "The Landscape Agency_sentence_6"
},
{
"operation": "Highlighting deleted",
"time": "79.95",
"value": "The Landscape Agency_sentence_5"
},
{
"operation": "Highlighting deleted",
"time": "81.867",
"value": "The Landscape Agency_cell_0_1_1"
},
{
"operation": "Highlighting deleted",
"time": "81.87",
"value": "The Landscape Agency_header_cell_0_1_0"
},
{
"operation": "Highlighting deleted",
"time": "82.462",
"value": "The Landscape Agency_cell_0_2_1"
},
{
"operation": "Highlighting deleted",
"time": "82.466",
"value": "The Landscape Agency_header_cell_0_2_0"
},
{
"operation": "finish",
"time": "144.77",
"value": "finish"
}
]
}
|
25121
|
Richard Kindleberger (June 17, 1942 – January 1, 2010 due to brain tumor) was an American newspaper reporter and editor from Baldwin, Maryland who worked at The Boston Globe.
|
{
"evidence": [
"Richard Kindleberger_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:\nRichard Kindleberger (June 17, 1942 – January 1, 2010 due to brain tumor) was an American newspaper reporter and editor from Baldwin, Maryland who worked at The Boston Globe.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Richard Kindleberger_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3010,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.088",
"value": "?search="
},
{
"operation": "search",
"time": "23.171",
"value": "Richard Kindleberger"
},
{
"operation": "Now on",
"time": "23.827",
"value": "Richard Kindleberger"
},
{
"operation": "Highlighting",
"time": "57.71",
"value": "Richard Kindleberger_sentence_14"
},
{
"operation": "Highlighting",
"time": "57.713",
"value": "Richard Kindleberger_section_2"
},
{
"operation": "Highlighting",
"time": "70.262",
"value": "Richard Kindleberger_sentence_0"
},
{
"operation": "Highlighting",
"time": "83.468",
"value": "Richard Kindleberger_sentence_1"
},
{
"operation": "finish",
"time": "102.973",
"value": "finish"
}
]
}
|
86824
|
The 1920 Clemson Tigers football team representing Clemson Agricultural College appeared in eleven matches, making a record of 6-4 and one tie.
|
{
"evidence": [
"1920 Clemson Tigers 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 1920 Clemson Tigers football team representing Clemson Agricultural College appeared in eleven matches, making a record of 6-4 and one tie.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"1920 Clemson Tigers football team_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3011,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.807",
"value": "undefined"
},
{
"operation": "search",
"time": "26.427",
"value": "1920 Clemson Tigers football team"
},
{
"operation": "Now on",
"time": "27.835",
"value": "1920 Clemson Tigers football team"
},
{
"operation": "Highlighting",
"time": "57.265",
"value": "1920 Clemson Tigers football team_sentence_0"
},
{
"operation": "Highlighting",
"time": "84.609",
"value": "1920 Clemson Tigers football team_cell_1_1_1"
},
{
"operation": "Highlighting",
"time": "84.615",
"value": "1920 Clemson Tigers football team_section_0"
},
{
"operation": "Highlighting",
"time": "84.619",
"value": "1920 Clemson Tigers football team_header_cell_1_0_1"
},
{
"operation": "Highlighting",
"time": "85.782",
"value": "1920 Clemson Tigers football team_cell_1_2_1"
},
{
"operation": "Highlighting",
"time": "87.317",
"value": "1920 Clemson Tigers football team_cell_1_3_1"
},
{
"operation": "Highlighting",
"time": "88.386",
"value": "1920 Clemson Tigers football team_cell_1_4_1"
},
{
"operation": "Highlighting",
"time": "89.336",
"value": "1920 Clemson Tigers football team_cell_1_5_1"
},
{
"operation": "Highlighting",
"time": "90.31",
"value": "1920 Clemson Tigers football team_cell_1_6_1"
},
{
"operation": "Highlighting",
"time": "91.382",
"value": "1920 Clemson Tigers football team_cell_1_7_1"
},
{
"operation": "Highlighting",
"time": "92.403",
"value": "1920 Clemson Tigers football team_cell_1_8_1"
},
{
"operation": "Highlighting",
"time": "93.363",
"value": "1920 Clemson Tigers football team_cell_1_9_1"
},
{
"operation": "Highlighting",
"time": "94.31",
"value": "1920 Clemson Tigers football team_cell_1_10_1"
},
{
"operation": "Highlighting",
"time": "95.318",
"value": "1920 Clemson Tigers football team_cell_1_11_1"
},
{
"operation": "Highlighting",
"time": "112.031",
"value": "1920 Clemson Tigers football team_cell_1_1_3"
},
{
"operation": "Highlighting",
"time": "112.036",
"value": "1920 Clemson Tigers football team_header_cell_1_0_3"
},
{
"operation": "Highlighting",
"time": "113.379",
"value": "1920 Clemson Tigers football team_cell_1_3_3"
},
{
"operation": "Highlighting",
"time": "114.375",
"value": "1920 Clemson Tigers football team_cell_1_4_3"
},
{
"operation": "Highlighting",
"time": "116.518",
"value": "1920 Clemson Tigers football team_cell_1_9_3"
},
{
"operation": "Highlighting",
"time": "136.442",
"value": "1920 Clemson Tigers football team_cell_1_5_3"
},
{
"operation": "Highlighting",
"time": "137.213",
"value": "1920 Clemson Tigers football team_cell_1_6_3"
},
{
"operation": "Highlighting",
"time": "138.027",
"value": "1920 Clemson Tigers football team_cell_1_7_3"
},
{
"operation": "Highlighting",
"time": "138.844",
"value": "1920 Clemson Tigers football team_cell_1_8_3"
},
{
"operation": "Highlighting",
"time": "140.059",
"value": "1920 Clemson Tigers football team_cell_1_10_3"
},
{
"operation": "Highlighting",
"time": "141.201",
"value": "1920 Clemson Tigers football team_cell_1_11_3"
},
{
"operation": "Highlighting",
"time": "148.254",
"value": "1920 Clemson Tigers football team_cell_1_2_3"
},
{
"operation": "Highlighting deleted",
"time": "151.38",
"value": "1920 Clemson Tigers football team_cell_1_1_1"
},
{
"operation": "Highlighting deleted",
"time": "152.164",
"value": "1920 Clemson Tigers football team_cell_1_2_1"
},
{
"operation": "Highlighting deleted",
"time": "153.149",
"value": "1920 Clemson Tigers football team_cell_1_3_1"
},
{
"operation": "Highlighting deleted",
"time": "153.983",
"value": "1920 Clemson Tigers football team_cell_1_4_1"
},
{
"operation": "Highlighting deleted",
"time": "157.834",
"value": "1920 Clemson Tigers football team_cell_1_11_1"
},
{
"operation": "Highlighting deleted",
"time": "158.717",
"value": "1920 Clemson Tigers football team_cell_1_10_1"
},
{
"operation": "Highlighting deleted",
"time": "159.776",
"value": "1920 Clemson Tigers football team_cell_1_9_1"
},
{
"operation": "Highlighting deleted",
"time": "160.654",
"value": "1920 Clemson Tigers football team_cell_1_8_1"
},
{
"operation": "Highlighting deleted",
"time": "161.604",
"value": "1920 Clemson Tigers football team_cell_1_7_1"
},
{
"operation": "Highlighting deleted",
"time": "162.462",
"value": "1920 Clemson Tigers football team_cell_1_6_1"
},
{
"operation": "Highlighting deleted",
"time": "163.443",
"value": "1920 Clemson Tigers football team_cell_1_5_1"
},
{
"operation": "Highlighting deleted",
"time": "163.449",
"value": "1920 Clemson Tigers football team_header_cell_1_0_1"
},
{
"operation": "Highlighting",
"time": "172.815",
"value": "1920 Clemson Tigers football team_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "182.435",
"value": "1920 Clemson Tigers football team_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "188.149",
"value": "1920 Clemson Tigers football team_cell_1_1_3"
},
{
"operation": "Highlighting deleted",
"time": "189.18",
"value": "1920 Clemson Tigers football team_cell_1_2_3"
},
{
"operation": "Highlighting deleted",
"time": "191.135",
"value": "1920 Clemson Tigers football team_cell_1_4_3"
},
{
"operation": "Highlighting deleted",
"time": "192.436",
"value": "1920 Clemson Tigers football team_cell_1_3_3"
},
{
"operation": "Highlighting deleted",
"time": "193.688",
"value": "1920 Clemson Tigers football team_cell_1_5_3"
},
{
"operation": "Highlighting deleted",
"time": "194.791",
"value": "1920 Clemson Tigers football team_cell_1_6_3"
},
{
"operation": "Highlighting deleted",
"time": "195.677",
"value": "1920 Clemson Tigers football team_cell_1_7_3"
},
{
"operation": "Highlighting deleted",
"time": "196.751",
"value": "1920 Clemson Tigers football team_cell_1_8_3"
},
{
"operation": "Highlighting deleted",
"time": "201.918",
"value": "1920 Clemson Tigers football team_cell_1_9_3"
},
{
"operation": "Highlighting deleted",
"time": "203.07",
"value": "1920 Clemson Tigers football team_cell_1_10_3"
},
{
"operation": "Highlighting deleted",
"time": "204.413",
"value": "1920 Clemson Tigers football team_cell_1_11_3"
},
{
"operation": "Highlighting deleted",
"time": "204.417",
"value": "1920 Clemson Tigers football team_section_0"
},
{
"operation": "Highlighting deleted",
"time": "204.421",
"value": "1920 Clemson Tigers football team_header_cell_1_0_3"
},
{
"operation": "finish",
"time": "232.325",
"value": "finish"
}
]
}
|
83066
|
Shefali Jariwala "Kaanta Laga Girl" (born 15 December 1982) is an Indian actress who debuted in 2002 and appeared in the television shows, Boogie Woogie(2008), Nach Baliye 5&7(2012-2016) and Bigg Boss 13 (2019-2020).
|
{
"evidence": [
"Shefali Jariwala_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:\nShefali Jariwala \"Kaanta Laga Girl\" (born 15 December 1982) is an Indian actress who debuted in 2002 and appeared in the television shows, Boogie Woogie(2008), Nach Baliye 5&7(2012-2016) and Bigg Boss 13 (2019-2020).\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Shefali Jariwala_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3012,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.117",
"value": "?search="
},
{
"operation": "search",
"time": "15.273",
"value": "Shefali Jariwala"
},
{
"operation": "Now on",
"time": "16.219",
"value": "Shefali Jariwala"
},
{
"operation": "Highlighting",
"time": "23.758",
"value": "Shefali Jariwala_sentence_0"
},
{
"operation": "finish",
"time": "43.282",
"value": "finish"
}
]
}
|
exfever_166
|
Doctor Doom was the first Marvel comic character to have a movie adaptation.
|
{
"evidence": [
"Doctor Doom",
"Comic book"
],
"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:\nDoctor Doom was the first Marvel comic character to have a movie adaptation.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Doctor Doom",
"Comic book"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3013,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
62817
|
Iraida Yusupova (born February 20, 1962) is a Turkmenistani composer of half Russian half Tatar ethnicity who lives in Moscow, Russia, Iraida Yusupova was born in Ashgabat, Turkmen SSR, and graduated from Moscow Conservatory with a degree in composition in 1987.
|
{
"evidence": [
"Iraida Yusupova_sentence_0",
"Iraida Yusupova_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:\nIraida Yusupova (born February 20, 1962) is a Turkmenistani composer of half Russian half Tatar ethnicity who lives in Moscow, Russia, Iraida Yusupova was born in Ashgabat, Turkmen SSR, and graduated from Moscow Conservatory with a degree in composition in 1987.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Iraida Yusupova_sentence_0",
"Iraida Yusupova_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3014,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.069",
"value": "?search="
},
{
"operation": "search",
"time": "26.066",
"value": "Iraida Yusupova"
},
{
"operation": "Now on",
"time": "26.916",
"value": "Iraida Yusupova"
},
{
"operation": "Highlighting",
"time": "169.795",
"value": "Iraida Yusupova_sentence_0"
},
{
"operation": "Highlighting",
"time": "180.365",
"value": "Iraida Yusupova_sentence_1"
},
{
"operation": "finish",
"time": "283.806",
"value": "finish"
}
]
}
|
71886
|
The California flood of 1605 affected California natives in industrial advancement populations, and the USGS revealed that the flood deposited a two-inch thick silt layer at the Santa Barbara basin making it the worst flood event in the past 2,000 years.
|
{
"evidence": [
"California flood of 1605_sentence_2"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe California flood of 1605 affected California natives in industrial advancement populations, and the USGS revealed that the flood deposited a two-inch thick silt layer at the Santa Barbara basin making it the worst flood event in the past 2,000 years.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"California flood of 1605_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3015,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.009",
"value": "?search="
},
{
"operation": "search",
"time": "17.588",
"value": "California flood of 1605"
},
{
"operation": "Now on",
"time": "19.257",
"value": "California flood of 1605"
},
{
"operation": "Highlighting",
"time": "26.112",
"value": "California flood of 1605_sentence_2"
},
{
"operation": "finish",
"time": "57.084",
"value": "finish"
}
]
}
|
exfever_1771
|
Bernard Madoff is the first fraudster to gain a Billionaire status.
|
{
"evidence": [
"Bernard Madoff",
"Chairman"
],
"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:\nBernard Madoff is the first fraudster to gain a Billionaire status.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Bernard Madoff",
"Chairman"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3016,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
94757
|
Cicely Courtneidge, born in 1893 remained single until passing away in Putney, in southwest London which had a population of 77,140 in 2011.
|
{
"evidence": [
"Cicely Courtneidge_sentence_4"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nCicely Courtneidge, born in 1893 remained single until passing away in Putney, in southwest London which had a population of 77,140 in 2011.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Cicely Courtneidge_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3017,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.649",
"value": "?search="
},
{
"operation": "search",
"time": "9.504",
"value": "Cicely Courtneidge"
},
{
"operation": "Now on",
"time": "10.573",
"value": "Cicely Courtneidge"
},
{
"operation": "Highlighting",
"time": "15.639",
"value": "Cicely Courtneidge_sentence_0"
},
{
"operation": "Highlighting",
"time": "28.885",
"value": "Cicely Courtneidge_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "41.071",
"value": "Cicely Courtneidge_sentence_0"
},
{
"operation": "finish",
"time": "46.781",
"value": "finish"
}
]
}
|
exfever_3175
|
Tom Hardy has since been excluded for an award presented annually by the Academy of Motion Picture Arts and Sciences.
|
{
"evidence": [
"Tom Hardy",
"Academy Award for Best Supporting Actor"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nTom Hardy has since been excluded for an award presented annually by the Academy of Motion Picture Arts and Sciences.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Tom Hardy",
"Academy Award for Best Supporting Actor"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3018,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3202
|
Sue Bird is an American-Israeli professional player in a team sport wherein a tied score at the end of regulation requires an additional period of play specified under the rules of the sport to bring a game to a decision.
|
{
"evidence": [
"Sue Bird",
"Basketball",
"Overtime (sports)"
],
"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:\nSue Bird is an American-Israeli professional player in a team sport wherein a tied score at the end of regulation requires an additional period of play specified under the rules of the sport to bring a game to a decision.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Sue Bird",
"Basketball",
"Overtime (sports)"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3019,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6256
|
A 1955 British CinemaScope adventure film directed by Terence Young starred an American actress, singer, dancer, and author that appeared in radio programs before her first formal foray into acting.
|
{
"evidence": [
"Janet Leigh",
"Safari (1956 film)",
"Janet Leigh"
],
"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:\nA 1955 British CinemaScope adventure film directed by Terence Young starred an American actress, singer, dancer, and author that appeared in radio programs before her first formal foray into acting.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Janet Leigh",
"Safari (1956 film)",
"Janet Leigh"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3020,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
65924
|
Afradapis is a genus of adapiform primate and is one of four typically European caenopithecines (a subfamily within the extinct primate family Adapidae) it also lost its anterior premolar
|
{
"evidence": [
"Afradapis_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:\nAfradapis is a genus of adapiform primate and is one of four typically European caenopithecines (a subfamily within the extinct primate family Adapidae) it also lost its anterior premolar\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Afradapis_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3021,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.02",
"value": "?search="
},
{
"operation": "search",
"time": "12.054",
"value": "Afradapis"
},
{
"operation": "Now on",
"time": "13.721",
"value": "Afradapis"
},
{
"operation": "Highlighting",
"time": "18.869",
"value": "Afradapis_sentence_0"
},
{
"operation": "Highlighting",
"time": "24.316",
"value": "Afradapis_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "25.252",
"value": "Afradapis_sentence_0"
},
{
"operation": "finish",
"time": "47.235",
"value": "finish"
}
]
}
|
88282
|
During the Chilean Civil War, the American government supported the forces of President Jose Manuel Balmaceda, causing troubled relations between the United States and the victorious congressionalist side; but just before the end of the Chilean Civil War, the United States sent USS Charleston to force the Chilean congressional cargo ship Itata to return to San Diego.
|
{
"evidence": [
"Baltimore crisis_sentence_12",
"Baltimore crisis_sentence_13",
"Baltimore crisis_sentence_14"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nDuring the Chilean Civil War, the American government supported the forces of President Jose Manuel Balmaceda, causing troubled relations between the United States and the victorious congressionalist side; but just before the end of the Chilean Civil War, the United States sent USS Charleston to force the Chilean congressional cargo ship Itata to return to San Diego.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Baltimore crisis_sentence_12",
"Baltimore crisis_sentence_13",
"Baltimore crisis_sentence_14"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3022,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.41",
"value": "?search="
},
{
"operation": "search",
"time": "40.247",
"value": "During the Chilean Civil War, the American government supported the forces of President Jose Manuel Balmaceda, causing troubled relations between the United States and"
},
{
"operation": "Now on",
"time": "43.333",
"value": "undefined"
},
{
"operation": "search",
"time": "139.92",
"value": "Baltimore crisis"
},
{
"operation": "Now on",
"time": "140.611",
"value": "Baltimore crisis"
},
{
"operation": "Page search",
"time": "168.334",
"value": "the American government supported the forces"
},
{
"operation": "Highlighting",
"time": "174.131",
"value": "Baltimore crisis_sentence_12"
},
{
"operation": "Highlighting",
"time": "186.01",
"value": "Baltimore crisis_sentence_13"
},
{
"operation": "Highlighting",
"time": "208.298",
"value": "Baltimore crisis_sentence_14"
},
{
"operation": "finish",
"time": "227.926",
"value": "finish"
}
]
}
|
59751
|
Anil Gurung, who was born and raised in Nepal, has scored multiple goals for his nation at the SAFF Championship throughout his career.
|
{
"evidence": [
"Anil Gurung_sentence_0",
"Anil Gurung_sentence_3",
"Anil Gurung_sentence_23"
],
"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:\nAnil Gurung, who was born and raised in Nepal, has scored multiple goals for his nation at the SAFF Championship throughout his career.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Anil Gurung_sentence_0",
"Anil Gurung_sentence_3",
"Anil Gurung_sentence_23"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3023,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.732",
"value": "?search="
},
{
"operation": "search",
"time": "34.258",
"value": "Anil Gurung"
},
{
"operation": "Now on",
"time": "35.431",
"value": "Anil Gurung"
},
{
"operation": "Page search",
"time": "43.592",
"value": "raise"
},
{
"operation": "Page search",
"time": "93.793",
"value": "nepal"
},
{
"operation": "Highlighting",
"time": "119.327",
"value": "Anil Gurung_sentence_0"
},
{
"operation": "Highlighting",
"time": "137.48",
"value": "Anil Gurung_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "160.835",
"value": "Anil Gurung_sentence_3"
},
{
"operation": "search",
"time": "170.596",
"value": "Barahi Secondary School"
},
{
"operation": "Now on",
"time": "171.831",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "179.098",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "179.293",
"value": "Anil Gurung"
},
{
"operation": "Highlighting",
"time": "187.939",
"value": "Anil Gurung_sentence_3"
},
{
"operation": "Page search",
"time": "234.827",
"value": "saff champion"
},
{
"operation": "Highlighting",
"time": "276.019",
"value": "Anil Gurung_sentence_23"
},
{
"operation": "hyperlink",
"time": "288.924",
"value": "2013 SAFF Championship"
},
{
"operation": "Now on",
"time": "289.8",
"value": "2013 SAFF Championship"
},
{
"operation": "Page search",
"time": "295.233",
"value": "gurung"
},
{
"operation": "hyperlink",
"time": "316.296",
"value": "Nepal"
},
{
"operation": "Now on",
"time": "319.161",
"value": "Nepal national football team"
},
{
"operation": "back-button-clicked",
"time": "321.005",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "321.191",
"value": "2013 SAFF Championship"
},
{
"operation": "finish",
"time": "334.455",
"value": "finish"
}
]
}
|
exfever_4815
|
Samsung Electronics, a subsidiary of the Samsung group, is also headquartered in Seoul, South Korea.
|
{
"evidence": [
"Samsung Life Insurance",
"Life insurance"
],
"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:\nSamsung Electronics, a subsidiary of the Samsung group, is also headquartered in Seoul, South Korea.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Samsung Life Insurance",
"Life insurance"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3024,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4312
|
In ancient Greek religion and mythology, Helios is the personification of the Sun who is considered a deity or a supernatural being who is divine and sacred.
|
{
"evidence": [
"Helios",
"Deity"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nIn ancient Greek religion and mythology, Helios is the personification of the Sun who is considered a deity or a supernatural being who is divine and sacred.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Helios",
"Deity"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3025,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5739
|
Tate Ellington is an American actor who has appeared in several television shows, including an American legal drama television series whose stars include an American actor.
|
{
"evidence": [
"Tate Ellington",
"The Good Wife",
"Josh Charles"
],
"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:\nTate Ellington is an American actor who has appeared in several television shows, including an American legal drama television series whose stars include an American actor.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Tate Ellington",
"The Good Wife",
"Josh Charles"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3026,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1670
|
Hearing loss may be caused by a number of factors, including a branch of biology concerned with the study of the differences in DNA among individuals.
|
{
"evidence": [
"Hearing loss",
"Genetics",
"Genetic variation"
],
"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:\nHearing loss may be caused by a number of factors, including a branch of biology concerned with the study of the differences in DNA among individuals.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Hearing loss",
"Genetics",
"Genetic variation"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3027,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5898
|
Brave is a 2015 American computer-animated?film genre that is considered a form of speculative fiction that seeks to elicit fear in its audience.
|
{
"evidence": [
"Brave (2012 film)",
"Fantasy film",
"Horror 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:\nBrave is a 2015 American computer-animated?film genre that is considered a form of speculative fiction that seeks to elicit fear in its audience.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Brave (2012 film)",
"Fantasy film",
"Horror film"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3028,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_668
|
Chet Atkin built the largest guitar museum in America.
|
{
"evidence": [
"Chet Atkins",
"George Barnes (musician)"
],
"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:\nChet Atkin built the largest guitar museum in America.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Chet Atkins",
"George Barnes (musician)"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3029,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2390
|
Rio is the highest-grossing 3D computer-animated by Blue Sky Studios.?
|
{
"evidence": [
"Rio (2011 film)",
"Jesse Eisenberg"
],
"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:\nRio is the highest-grossing 3D computer-animated by Blue Sky Studios.?\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Rio (2011 film)",
"Jesse Eisenberg"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3030,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
36714
|
Welcome to the Morbid Reich is the first album by the Polish band Vader and received favorable ratings from numerous sources.
|
{
"evidence": [
"Welcome to the Morbid Reich_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:\nWelcome to the Morbid Reich is the first album by the Polish band Vader and received favorable ratings from numerous sources.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Welcome to the Morbid Reich_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3031,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.745",
"value": "?search="
},
{
"operation": "search",
"time": "13.23",
"value": "Welcome to the Morbid Reich"
},
{
"operation": "Now on",
"time": "14.501",
"value": "Welcome to the Morbid Reich"
},
{
"operation": "Highlighting",
"time": "20.692",
"value": "Welcome to the Morbid Reich_sentence_0"
},
{
"operation": "finish",
"time": "30.784",
"value": "finish"
}
]
}
|
7201
|
The Front of the Bolivian Revolution, established in 1965 for the 1966 presidential and congressional elections, presented as its presidential candidate Luis Adolfo Siles Salinas (independent), a Bolivian military officer and politician who served as the 47th President of Bolivia twice non-consecutively from 1964 to 1966 and from 1966 to 1969, and Luis Adolfo Siles Salinas (PSD), as vice-presidential candidate.
|
{
"evidence": [
"Front of the Bolivian Revolution_sentence_4"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Front of the Bolivian Revolution, established in 1965 for the 1966 presidential and congressional elections, presented as its presidential candidate Luis Adolfo Siles Salinas (independent), a Bolivian military officer and politician who served as the 47th President of Bolivia twice non-consecutively from 1964 to 1966 and from 1966 to 1969, and Luis Adolfo Siles Salinas (PSD), as vice-presidential candidate.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Front of the Bolivian Revolution_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3032,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.709",
"value": "?search="
},
{
"operation": "search",
"time": "14.282",
"value": "Front of the Bolivian Revolution"
},
{
"operation": "Now on",
"time": "15.2",
"value": "Front of the Bolivian Revolution"
},
{
"operation": "Highlighting",
"time": "58.518",
"value": "Front of the Bolivian Revolution_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "100.044",
"value": "Front of the Bolivian Revolution_sentence_0"
},
{
"operation": "Highlighting",
"time": "104.523",
"value": "Front of the Bolivian Revolution_sentence_3"
},
{
"operation": "Highlighting",
"time": "111.481",
"value": "Front of the Bolivian Revolution_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "234.202",
"value": "Front of the Bolivian Revolution_sentence_3"
},
{
"operation": "finish",
"time": "242.931",
"value": "finish"
}
]
}
|
27851
|
The majority of Cordillera voters did not want it to be an autonomous region because of the effect on their families and the possibility of terrorism; none of the provinces voted for autonomy.
|
{
"evidence": [
"Cordillera autonomy movement_sentence_25",
"Cordillera autonomy movement_sentence_27",
"Cordillera autonomy movement_sentence_28",
"Cordillera autonomy movement_sentence_29",
"Cordillera autonomy movement_sentence_30",
"Cordillera autonomy movement_sentence_31"
],
"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 majority of Cordillera voters did not want it to be an autonomous region because of the effect on their families and the possibility of terrorism; none of the provinces voted for autonomy.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Cordillera autonomy movement_sentence_25",
"Cordillera autonomy movement_sentence_27",
"Cordillera autonomy movement_sentence_28",
"Cordillera autonomy movement_sentence_29",
"Cordillera autonomy movement_sentence_30",
"Cordillera autonomy movement_sentence_31"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3033,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.107",
"value": "?search="
},
{
"operation": "search",
"time": "16.123",
"value": "Cordillera"
},
{
"operation": "Now on",
"time": "16.816",
"value": "Cordillera"
},
{
"operation": "search",
"time": "26.75",
"value": "Cordillera Administrative Region"
},
{
"operation": "Now on",
"time": "28.161",
"value": "Cordillera Administrative Region"
},
{
"operation": "hyperlink",
"time": "88.67",
"value": "autonomy in the Cordillera region"
},
{
"operation": "Now on",
"time": "89.204",
"value": "Cordillera autonomy movement"
},
{
"operation": "hyperlink",
"time": "163.061",
"value": "Cordilleras"
},
{
"operation": "Now on",
"time": "163.701",
"value": "Cordillera Administrative Region"
},
{
"operation": "hyperlink",
"time": "219.965",
"value": "Cordillera autonomy movement"
},
{
"operation": "Now on",
"time": "220.328",
"value": "Cordillera autonomy movement"
},
{
"operation": "Highlighting",
"time": "264.128",
"value": "Cordillera autonomy movement_sentence_25"
},
{
"operation": "Highlighting",
"time": "269.367",
"value": "Cordillera autonomy movement_sentence_28"
},
{
"operation": "Highlighting deleted",
"time": "271.191",
"value": "Cordillera autonomy movement_sentence_28"
},
{
"operation": "Highlighting",
"time": "271.813",
"value": "Cordillera autonomy movement_sentence_27"
},
{
"operation": "Highlighting",
"time": "272.526",
"value": "Cordillera autonomy movement_sentence_28"
},
{
"operation": "Highlighting",
"time": "275.301",
"value": "Cordillera autonomy movement_sentence_29"
},
{
"operation": "Highlighting",
"time": "277.405",
"value": "Cordillera autonomy movement_sentence_30"
},
{
"operation": "Highlighting",
"time": "287.598",
"value": "Cordillera autonomy movement_sentence_31"
},
{
"operation": "finish",
"time": "314.727",
"value": "finish"
}
]
}
|
exfever_4031
|
The Houston Rockets are an American professional basketball team based in Houston that was established in a city in the U.S. state of California on the coast of the Pacific Ocean in 1968.
|
{
"evidence": [
"Houston Rockets",
"San Diego"
],
"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 Houston Rockets are an American professional basketball team based in Houston that was established in a city in the U.S. state of California on the coast of the Pacific Ocean in 1968.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Houston Rockets",
"San Diego"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3034,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4579
|
The Craft is a 1996 Canadian supernatural horror teen film that stars an American film and television actress and former fashion model, known for her role as Mona Thorne in an American broadcast television network that was originally owned by Chris-Craft Industries' United Television.
|
{
"evidence": [
"The Craft (film)",
"Rachel True",
"UPN"
],
"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 Craft is a 1996 Canadian supernatural horror teen film that stars an American film and television actress and former fashion model, known for her role as Mona Thorne in an American broadcast television network that was originally owned by Chris-Craft Industries' United Television.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Craft (film)",
"Rachel True",
"UPN"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3035,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
38624
|
Currently the largest pizza restaurant chain in Thailand, The Pizza Company began to expand in 2004 and franchise internationally and today the restaurant has franchises in Jordan, the United Arab Emirates, Saudi Arabia, Laos, China, Bahrain, Myanmar and Cambodia.
|
{
"evidence": [
"The Pizza Company_sentence_8",
"The Pizza Company_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:\nCurrently the largest pizza restaurant chain in Thailand, The Pizza Company began to expand in 2004 and franchise internationally and today the restaurant has franchises in Jordan, the United Arab Emirates, Saudi Arabia, Laos, China, Bahrain, Myanmar and Cambodia.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"The Pizza Company_sentence_8",
"The Pizza Company_sentence_6"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3036,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.72",
"value": "?search="
},
{
"operation": "search",
"time": "18.219",
"value": "The Pizza Company"
},
{
"operation": "Now on",
"time": "19.584",
"value": "The Pizza Company"
},
{
"operation": "Page search",
"time": "52.897",
"value": "larg"
},
{
"operation": "Highlighting",
"time": "62.2",
"value": "The Pizza Company_sentence_8"
},
{
"operation": "Highlighting",
"time": "72.203",
"value": "The Pizza Company_sentence_6"
},
{
"operation": "finish",
"time": "111.662",
"value": "finish"
}
]
}
|
exfever_2194
|
Red Hot Chili Peppers' eleventh studio is The Getaway following the tenth studio album that was released by Warner Bros. Records in 2011.
|
{
"evidence": [
"The Getaway (Red Hot Chili Peppers album)",
"I'm with You (album)"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nRed Hot Chili Peppers' eleventh studio is The Getaway following the tenth studio album that was released by Warner Bros. Records in 2011.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Getaway (Red Hot Chili Peppers album)",
"I'm with You (album)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3037,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
4253
|
John Winans (Born September 27, 1831) was a candidate of the General election, November 4, 1873 and lost with 741 votes.
|
{
"evidence": [
"John Winans_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:\nJohn Winans (Born September 27, 1831) was a candidate of the General election, November 4, 1873 and lost with 741 votes.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"John Winans_sentence_9"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3038,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "search",
"time": "19.255",
"value": "John Winans"
},
{
"operation": "Highlighting",
"time": "24.585",
"value": "John Winans_sentence_0"
},
{
"operation": "Highlighting",
"time": "90.756",
"value": "John Winans_sentence_8"
},
{
"operation": "Highlighting",
"time": "93.152",
"value": "John Winans_sentence_9"
},
{
"operation": "Highlighting deleted",
"time": "158.104",
"value": "John Winans_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "170.613",
"value": "John Winans_sentence_8"
},
{
"operation": "finish",
"time": "172.97",
"value": "finish"
}
]
}
|
74055
|
Joe Magidsohn scored one touchdown with a total of ten points, while Frederick L. Conklin scored three extra points and one field goal with a total of six points.
|
{
"evidence": [
"Joe Magidsohn_sentence_0",
"Joe Magidsohn_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:\nJoe Magidsohn scored one touchdown with a total of ten points, while Frederick L. Conklin scored three extra points and one field goal with a total of six points.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Joe Magidsohn_sentence_0",
"Joe Magidsohn_sentence_1"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3039,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.099",
"value": "?search="
},
{
"operation": "search",
"time": "12.901",
"value": "Joe Magidsohn"
},
{
"operation": "Now on",
"time": "13.793",
"value": "Joe Magidsohn"
},
{
"operation": "Highlighting",
"time": "158.996",
"value": "Joe Magidsohn_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "160.66",
"value": "Joe Magidsohn_sentence_0"
},
{
"operation": "Page search",
"time": "184.789",
"value": "frederick"
},
{
"operation": "hyperlink",
"time": "202.508",
"value": "University of Michigan"
},
{
"operation": "Now on",
"time": "207.533",
"value": "University of Michigan"
},
{
"operation": "search",
"time": "230.235",
"value": "1909 University of Michigan football"
},
{
"operation": "Now on",
"time": "233.412",
"value": "undefined"
},
{
"operation": "search",
"time": "249.311",
"value": "1909 college football season"
},
{
"operation": "Now on",
"time": "252.401",
"value": "1909 college football season"
},
{
"operation": "Page search",
"time": "261.784",
"value": "magidoshn"
},
{
"operation": "search",
"time": "279.274",
"value": "frederick l conklin"
},
{
"operation": "Now on",
"time": "280.64",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "290.533",
"value": "1911 Michigan Wolverines football team"
},
{
"operation": "Now on",
"time": "291.478",
"value": "1911 Michigan Wolverines football team"
},
{
"operation": "Page search",
"time": "388.68",
"value": "magidshon"
},
{
"operation": "Page search",
"time": "398.358",
"value": "magidsohn"
},
{
"operation": "search",
"time": "436.592",
"value": "Joe Magidsohn"
},
{
"operation": "Now on",
"time": "437.397",
"value": "Joe Magidsohn"
},
{
"operation": "Highlighting",
"time": "464.189",
"value": "Joe Magidsohn_sentence_0"
},
{
"operation": "Highlighting",
"time": "466.996",
"value": "Joe Magidsohn_sentence_1"
},
{
"operation": "Page search",
"time": "476.5",
"value": "field goal"
},
{
"operation": "Page search",
"time": "482.909",
"value": "touchdown"
},
{
"operation": "finish",
"time": "506.26",
"value": "finish"
}
]
}
|
exfever_4932
|
Samuel L. Jackson rose to fame with films including a 1993 American romantic black comedy crime film directed by Tony Scott.
|
{
"evidence": [
"Samuel L. Jackson",
"True Romance"
],
"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:\nSamuel L. Jackson rose to fame with films including a 1993 American romantic black comedy crime film directed by Tony Scott.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Samuel L. Jackson",
"True Romance"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3040,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
21147
|
Barangaroo ferry wharf was the only wharf to service Darling Harbour until Pyrmont Bay was opened in the early 2000s and the wharf was to be situated at the Darling Harbour wharf, and services to the Aquarium wharf, 250 metres to the east, phased out.
|
{
"evidence": [
"Barangaroo ferry wharf_sentence_9",
"Barangaroo ferry wharf_sentence_11"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBarangaroo ferry wharf was the only wharf to service Darling Harbour until Pyrmont Bay was opened in the early 2000s and the wharf was to be situated at the Darling Harbour wharf, and services to the Aquarium wharf, 250 metres to the east, phased out.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Barangaroo ferry wharf_sentence_9",
"Barangaroo ferry wharf_sentence_11"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3041,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.521",
"value": "?search="
},
{
"operation": "search",
"time": "9.193",
"value": "Barangaroo ferry wharf"
},
{
"operation": "Now on",
"time": "9.878",
"value": "Barangaroo ferry wharf"
},
{
"operation": "Page search",
"time": "20.298",
"value": "Darling Harbou"
},
{
"operation": "Highlighting",
"time": "43.589",
"value": "Barangaroo ferry wharf_sentence_9"
},
{
"operation": "Highlighting",
"time": "57.254",
"value": "Barangaroo ferry wharf_sentence_11"
},
{
"operation": "finish",
"time": "82.921",
"value": "finish"
}
]
}
|
63519
|
Paudge Behan was the son of the Chief of Staff of the IRA, which fought a guerilla war against Britain.
|
{
"evidence": [
"Paudge Behan_sentence_1",
"Irish Republican Army (1922–1969)_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:\nPaudge Behan was the son of the Chief of Staff of the IRA, which fought a guerilla war against Britain.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Paudge Behan_sentence_1",
"Irish Republican Army (1922–1969)_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3042,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.435",
"value": "?search="
},
{
"operation": "search",
"time": "68.757",
"value": "Paudge Behan"
},
{
"operation": "Now on",
"time": "70.018",
"value": "Paudge Behan"
},
{
"operation": "Highlighting",
"time": "80.358",
"value": "Paudge Behan_sentence_1"
},
{
"operation": "hyperlink",
"time": "89.436",
"value": "Chief of Staff"
},
{
"operation": "Now on",
"time": "91.1",
"value": "List of IRA Chiefs of Staff"
},
{
"operation": "back-button-clicked",
"time": "113.052",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "113.213",
"value": "Paudge Behan"
},
{
"operation": "hyperlink",
"time": "116.699",
"value": "IRA"
},
{
"operation": "Now on",
"time": "117.79",
"value": "Irish Republican Army (1922–1969)"
},
{
"operation": "Highlighting",
"time": "120.974",
"value": "Irish Republican Army (1922–1969)_sentence_0"
},
{
"operation": "finish",
"time": "135.771",
"value": "finish"
}
]
}
|
35944
|
In 1892, Grantown-on-Spey in Strathspeym formed the first shinty club, a club that plays shinty/hurling that appears prominently in the legend of Cúchulainn, the Celtic mythology, and during the late nineteenth century, twelve shinty clubs from across the Badenoch and Strathspey area were in operation.
|
{
"evidence": [
"Strathspey Camanachd_sentence_4",
"Shinty_sentence_41",
"Shinty_sentence_13",
"Grantown-on-Spey_sentence_0",
"Nethy Bridge_sentence_0",
"Boat of Garten_sentence_0",
"Dulnain Bridge_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:\nIn 1892, Grantown-on-Spey in Strathspeym formed the first shinty club, a club that plays shinty/hurling that appears prominently in the legend of Cúchulainn, the Celtic mythology, and during the late nineteenth century, twelve shinty clubs from across the Badenoch and Strathspey area were in operation.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Strathspey Camanachd_sentence_4",
"Shinty_sentence_41",
"Shinty_sentence_13",
"Grantown-on-Spey_sentence_0",
"Nethy Bridge_sentence_0",
"Boat of Garten_sentence_0",
"Dulnain Bridge_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3043,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.621",
"value": "?search="
},
{
"operation": "search",
"time": "31.102",
"value": "Contains...shinty club"
},
{
"operation": "Now on",
"time": "32.069",
"value": "undefined"
},
{
"operation": "search",
"time": "63.717",
"value": "Grantown-on-Spey"
},
{
"operation": "Now on",
"time": "64.431",
"value": "Grantown-on-Spey"
},
{
"operation": "search",
"time": "88.044",
"value": "first shinty club"
},
{
"operation": "Now on",
"time": "89.135",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "105.544",
"value": "History of Shinty in Strathspey Strathspey Camanachd_section_0"
},
{
"operation": "Now on",
"time": "106.689",
"value": "Strathspey Camanachd#History of Shinty in Strathspey Strathspey Camanachd section 0"
},
{
"operation": "hyperlink",
"time": "167.285",
"value": "shinty"
},
{
"operation": "Now on",
"time": "167.926",
"value": "Shinty"
},
{
"operation": "Highlighting",
"time": "205.468",
"value": "Shinty_sentence_41"
},
{
"operation": "search",
"time": "268.468",
"value": "Grantown-on-Spey"
},
{
"operation": "Now on",
"time": "269.189",
"value": "Grantown-on-Spey"
},
{
"operation": "search",
"time": "301.886",
"value": "Contains...Strathspeym"
},
{
"operation": "Now on",
"time": "302.916",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "314.06",
"value": "Strathspey"
},
{
"operation": "Now on",
"time": "314.799",
"value": "Strathspey"
},
{
"operation": "hyperlink",
"time": "328.725",
"value": "Strathspey Camanachd"
},
{
"operation": "Now on",
"time": "329.161",
"value": "Strathspey Camanachd"
},
{
"operation": "Highlighting",
"time": "343.246",
"value": "Strathspey Camanachd_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "429.47",
"value": "Strathspey Camanachd_sentence_1"
},
{
"operation": "Highlighting",
"time": "440.146",
"value": "Strathspey Camanachd_sentence_4"
},
{
"operation": "search",
"time": "498.229",
"value": "Shinty"
},
{
"operation": "Now on",
"time": "499.177",
"value": "Shinty"
},
{
"operation": "Highlighting",
"time": "576.713",
"value": "Shinty_sentence_41"
},
{
"operation": "Highlighting",
"time": "595.93",
"value": "Shinty_sentence_13"
},
{
"operation": "Page search",
"time": "638.936",
"value": "Badenoch"
},
{
"operation": "hyperlink",
"time": "707.545",
"value": "Shinty league system"
},
{
"operation": "Now on",
"time": "708.065",
"value": "Shinty league system"
},
{
"operation": "hyperlink",
"time": "744.565",
"value": "shinty"
},
{
"operation": "Now on",
"time": "745.037",
"value": "Shinty"
},
{
"operation": "search",
"time": "774.973",
"value": "Grantown-on-Spey"
},
{
"operation": "Now on",
"time": "775.696",
"value": "Grantown-on-Spey"
},
{
"operation": "Highlighting",
"time": "800.334",
"value": "Grantown-on-Spey_sentence_0"
},
{
"operation": "search",
"time": "811.054",
"value": "Strathspey"
},
{
"operation": "Now on",
"time": "811.958",
"value": "Strathspey"
},
{
"operation": "search",
"time": "850.949",
"value": "nineteenth century shinty"
},
{
"operation": "Now on",
"time": "852.167",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "863.558",
"value": "History of Shinty in Strathspey Strathspey Camanachd_section_0"
},
{
"operation": "Now on",
"time": "864.304",
"value": "Strathspey Camanachd#History of Shinty in Strathspey Strathspey Camanachd section 0"
},
{
"operation": "hyperlink",
"time": "902.648",
"value": "Nethy Bridge"
},
{
"operation": "Now on",
"time": "903.17",
"value": "Nethy Bridge"
},
{
"operation": "Highlighting",
"time": "908.386",
"value": "Nethy Bridge_sentence_0"
},
{
"operation": "back-button-clicked",
"time": "910.524",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "910.66",
"value": "Strathspey Camanachd#History of Shinty in Strathspey Strathspey Camanachd section 0"
},
{
"operation": "hyperlink",
"time": "916.445",
"value": "Boat of Garten"
},
{
"operation": "Now on",
"time": "916.993",
"value": "Boat of Garten"
},
{
"operation": "Highlighting",
"time": "919.543",
"value": "Boat of Garten_sentence_0"
},
{
"operation": "back-button-clicked",
"time": "921.76",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "921.895",
"value": "Strathspey Camanachd#History of Shinty in Strathspey Strathspey Camanachd section 0"
},
{
"operation": "hyperlink",
"time": "926.151",
"value": "Dulnain Bridge"
},
{
"operation": "Now on",
"time": "926.84",
"value": "Dulnain Bridge"
},
{
"operation": "Highlighting",
"time": "934.175",
"value": "Dulnain Bridge_sentence_0"
},
{
"operation": "finish",
"time": "951.165",
"value": "finish"
}
]
}
|
exfever_6289
|
Lorelai Gilmore is a character in the WB dramedy television series Gilmore Girls,?portrayed by an American actress who received nominations for the Screen Actors Guild.
|
{
"evidence": [
"Lorelai Gilmore",
"Lauren Graham"
],
"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:\nLorelai Gilmore is a character in the WB dramedy television series Gilmore Girls,?portrayed by an American actress who received nominations for the Screen Actors Guild.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Lorelai Gilmore",
"Lauren Graham"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3044,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3334
|
Lucas is a 1986 American romantic comedy-drama film directed by David Seltzer that stars an American actor who has appeared in films, including Platoon.
|
{
"evidence": [
"Lucas (film)",
"Charlie Sheen"
],
"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:\nLucas is a 1986 American romantic comedy-drama film directed by David Seltzer that stars an American actor who has appeared in films, including Platoon.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Lucas (film)",
"Charlie Sheen"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3045,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6621
|
Jean-Claude Van Damme is a Canadian actor, martial artist, filmmaker, and fight choreographer who became a popular action film star and followed up with a 1990 film directed by Sheldon Lettich.
|
{
"evidence": [
"Jean-Claude Van Damme",
"Lionheart (1990 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:\nJean-Claude Van Damme is a Canadian actor, martial artist, filmmaker, and fight choreographer who became a popular action film star and followed up with a 1990 film directed by Sheldon Lettich.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Jean-Claude Van Damme",
"Lionheart (1990 film)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3046,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
59945
|
The United States Ambassador to Luxembourg supervises the embassy staff, the embassy staff provides consular services and from 1903 to 1923 the Ambassador to the Netherlands served concurrently as Ambassador to Luxembourg.
|
{
"evidence": [
"List of ambassadors of the United States to Luxembourg_sentence_0",
"List of ambassadors of the United States to Luxembourg_sentence_2",
"List of ambassadors of the United States to Luxembourg_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:\nThe United States Ambassador to Luxembourg supervises the embassy staff, the embassy staff provides consular services and from 1903 to 1923 the Ambassador to the Netherlands served concurrently as Ambassador to Luxembourg.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"List of ambassadors of the United States to Luxembourg_sentence_0",
"List of ambassadors of the United States to Luxembourg_sentence_2",
"List of ambassadors of the United States to Luxembourg_sentence_4"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3047,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.237",
"value": "?search="
},
{
"operation": "search",
"time": "21.742",
"value": "United States Ambassador to Luxembourg"
},
{
"operation": "Now on",
"time": "23.345",
"value": "List of ambassadors of the United States to Luxembourg"
},
{
"operation": "Highlighting",
"time": "29.285",
"value": "List of ambassadors of the United States to Luxembourg_sentence_0"
},
{
"operation": "Highlighting",
"time": "30.518",
"value": "List of ambassadors of the United States to Luxembourg_sentence_1"
},
{
"operation": "Highlighting",
"time": "47.132",
"value": "List of ambassadors of the United States to Luxembourg_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "48.19",
"value": "List of ambassadors of the United States to Luxembourg_sentence_1"
},
{
"operation": "Highlighting",
"time": "93.262",
"value": "List of ambassadors of the United States to Luxembourg_sentence_4"
},
{
"operation": "finish",
"time": "104.577",
"value": "finish"
}
]
}
|
42003
|
The General Coordinating Committee of the Ministerial Conference on the Protection of Forests in Europe has five representatives from the signatory countries and serves as the financing body for the operation of its Liaison Unit Bratislava, currently based in Slovakia.
|
{
"evidence": [
"Ministerial Conference on the Protection of Forests in Europe_sentence_42"
],
"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 General Coordinating Committee of the Ministerial Conference on the Protection of Forests in Europe has five representatives from the signatory countries and serves as the financing body for the operation of its Liaison Unit Bratislava, currently based in Slovakia.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Ministerial Conference on the Protection of Forests in Europe_sentence_42"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3048,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.001",
"value": "?search="
},
{
"operation": "search",
"time": "0.009",
"value": "General Coordinating Committee of the Ministerial Conference on the Protection of Forests in Europe"
},
{
"operation": "Now on",
"time": "0.011",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "0.028",
"value": "Ministerial Conference on the Protection of Forests in Europe"
},
{
"operation": "Now on",
"time": "0.028",
"value": "Ministerial Conference on the Protection of Forests in Europe"
},
{
"operation": "Highlighting",
"time": "0.041",
"value": "Ministerial Conference on the Protection of Forests in Europe_sentence_42"
},
{
"operation": "finish",
"time": "0.311",
"value": "finish"
},
{
"operation": "Now on",
"time": "1620744052.252",
"value": "Ministerial Conference on the Protection of Forests in Europe"
},
{
"operation": "finish",
"time": "1620744054.779",
"value": "finish"
}
]
}
|
exfever_169
|
The Twilight Zone is an American media franchise based on the anthology which can be broadcast via over-the-air, satellite, or cable.
|
{
"evidence": [
"The Twilight Zone",
"Television program"
],
"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 Twilight Zone is an American media franchise based on the anthology which can be broadcast via over-the-air, satellite, or cable.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Twilight Zone",
"Television program"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3049,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
10652
|
The company, which was listed as 189 on the magazine Inc. and achieved more than 2000% growth rate in three years of operation, is EmazingLights.
|
{
"evidence": [
"EmazingLights_sentence_2",
"EmazingLights_sentence_3"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe company, which was listed as 189 on the magazine Inc. and achieved more than 2000% growth rate in three years of operation, is EmazingLights.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"EmazingLights_sentence_2",
"EmazingLights_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3050,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.064",
"value": "?search="
},
{
"operation": "search",
"time": "9.887",
"value": "emazinglights"
},
{
"operation": "Now on",
"time": "10.672",
"value": "EmazingLights"
},
{
"operation": "Highlighting",
"time": "180.917",
"value": "EmazingLights_sentence_2"
},
{
"operation": "Highlighting",
"time": "181.873",
"value": "EmazingLights_sentence_3"
},
{
"operation": "finish",
"time": "197.557",
"value": "finish"
}
]
}
|
exfever_4523
|
Conan graduated from Harvard University as Cum Laude.
|
{
"evidence": [
"Conan O'Brien",
"The Simpsons"
],
"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:\nConan graduated from Harvard University as Cum Laude.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Conan O'Brien",
"The Simpsons"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3051,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
21280
|
A genus of flowering plants belong to a distinct kingdom, order and tribe.
|
{
"evidence": [
"Flowering plant_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:\nA genus of flowering plants belong to a distinct kingdom, order and tribe.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Flowering plant_sentence_2"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3052,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.571",
"value": "?search="
},
{
"operation": "search",
"time": "35.762",
"value": "a genus of flowering plants belong to a distinct kingdom, order and tribe"
},
{
"operation": "Now on",
"time": "38.09",
"value": "undefined"
},
{
"operation": "search",
"time": "52.929",
"value": "Contains...\"distinct kingdom, order, and tribe\""
},
{
"operation": "Now on",
"time": "53.484",
"value": "undefined"
},
{
"operation": "search",
"time": "62.538",
"value": "Contains...\"a genus of flowering plants\""
},
{
"operation": "Now on",
"time": "64.635",
"value": "undefined"
},
{
"operation": "search",
"time": "93.702",
"value": "Contains...\"a genus of flowering plants belongs\""
},
{
"operation": "Now on",
"time": "94.088",
"value": "undefined"
},
{
"operation": "search",
"time": "105.449",
"value": "Contains...\"distinct kingdom\""
},
{
"operation": "Now on",
"time": "106.043",
"value": "undefined"
},
{
"operation": "search",
"time": "114.269",
"value": "Contains... a genus of flowering plants"
},
{
"operation": "Now on",
"time": "115.289",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "116.095",
"value": "Aechmea"
},
{
"operation": "Now on",
"time": "116.902",
"value": "Aechmea"
},
{
"operation": "Page search",
"time": "135.288",
"value": "tribe"
},
{
"operation": "search",
"time": "152.699",
"value": "Contains... a genus of flowering plants"
},
{
"operation": "Now on",
"time": "153.448",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "156.085",
"value": "Schefflera"
},
{
"operation": "Now on",
"time": "156.911",
"value": "Schefflera"
},
{
"operation": "search",
"time": "193.654",
"value": "Contains... a genus of flowering plants"
},
{
"operation": "Now on",
"time": "194.884",
"value": "undefined"
},
{
"operation": "search",
"time": "223.52",
"value": "Flowering plants"
},
{
"operation": "Now on",
"time": "224.958",
"value": "Flowering plants"
},
{
"operation": "Highlighting",
"time": "237.22",
"value": "Flowering plant_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "237.876",
"value": "Flowering plant_sentence_1"
},
{
"operation": "Highlighting",
"time": "239.152",
"value": "Flowering plant_sentence_2"
},
{
"operation": "Page search",
"time": "255.352",
"value": "kingdom order"
},
{
"operation": "Page search",
"time": "264.782",
"value": "order"
},
{
"operation": "Highlighting deleted",
"time": "325.655",
"value": "Flowering plant_sentence_2"
},
{
"operation": "Highlighting",
"time": "326.334",
"value": "Flowering plant_sentence_2"
},
{
"operation": "Page search",
"time": "350.466",
"value": "distinct"
},
{
"operation": "page-search-reset",
"time": "376.428",
"value": "page-search-reset"
},
{
"operation": "Highlighting deleted",
"time": "378.305",
"value": "Flowering plant_sentence_2"
},
{
"operation": "Highlighting",
"time": "378.866",
"value": "Flowering plant_sentence_2"
},
{
"operation": "hyperlink",
"time": "414.114",
"value": "genera"
},
{
"operation": "Now on",
"time": "414.919",
"value": "Genera"
},
{
"operation": "Page search",
"time": "420.623",
"value": "distinct"
},
{
"operation": "Page search",
"time": "436.798",
"value": "order"
},
{
"operation": "Page search",
"time": "471.439",
"value": "plants"
},
{
"operation": "hyperlink",
"time": "491.365",
"value": "List of the largest genera of flowering plants"
},
{
"operation": "Now on",
"time": "492.955",
"value": "List of the largest genera of flowering plants"
},
{
"operation": "Page search",
"time": "499.65",
"value": "distinct"
},
{
"operation": "search",
"time": "543.073",
"value": "Flowering plant"
},
{
"operation": "Now on",
"time": "544.84",
"value": "Flowering plant"
},
{
"operation": "search",
"time": "558.232",
"value": "Contains...genera of flowering plants"
},
{
"operation": "Now on",
"time": "559.941",
"value": "undefined"
},
{
"operation": "finish",
"time": "631.507",
"value": "finish"
}
]
}
|
exfever_3565
|
This Is Us will be made into a film in 2023.
|
{
"evidence": [
"This Is Us (TV series)",
"Justin Hartley"
],
"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:\nThis Is Us will be made into a film in 2023.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"This Is Us (TV series)",
"Justin Hartley"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3053,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
74198
|
Larry Ross was third in the British League World Riders Championship in 1980, and rode Long Track the next year; from 1977 to 1986, he rode in eight finals of the World Pairs and was runnerup with Ivan Mauger in 1978 and 1981.
|
{
"evidence": [
"Larry Ross_sentence_6"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nLarry Ross was third in the British League World Riders Championship in 1980, and rode Long Track the next year; from 1977 to 1986, he rode in eight finals of the World Pairs and was runnerup with Ivan Mauger in 1978 and 1981.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Larry Ross_sentence_6"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3054,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.261",
"value": "?search="
},
{
"operation": "search",
"time": "226.265",
"value": "Larry Ross"
},
{
"operation": "Now on",
"time": "227.251",
"value": "Larry Ross"
},
{
"operation": "Highlighting",
"time": "342.077",
"value": "Larry Ross_sentence_6"
},
{
"operation": "Highlighting",
"time": "443.131",
"value": "Larry Ross_sentence_7"
},
{
"operation": "Highlighting",
"time": "464.755",
"value": "Larry Ross_sentence_8"
},
{
"operation": "Highlighting deleted",
"time": "834.482",
"value": "Larry Ross_sentence_8"
},
{
"operation": "Highlighting deleted",
"time": "838.858",
"value": "Larry Ross_sentence_7"
},
{
"operation": "finish",
"time": "980.506",
"value": "finish"
}
]
}
|
exfever_4509
|
STS-96 was a Space Shuttle mission to a modular space station that is a multinational collaborative project of space agencies such as an independent agency of the U.S. federal government responsible for the civilian space program, as well as aeronautics and space research.
|
{
"evidence": [
"STS-96",
"International Space Station",
"NASA"
],
"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:\nSTS-96 was a Space Shuttle mission to a modular space station that is a multinational collaborative project of space agencies such as an independent agency of the U.S. federal government responsible for the civilian space program, as well as aeronautics and space research.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"STS-96",
"International Space Station",
"NASA"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3055,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
17753
|
The District of Columbia Organic Act of 1871 allowed the cities of Washington, Alexandria and Georgetown to each maintain their own municipal governments but Alexandria County was returned by Congress to the state of Virginia in 1846.
|
{
"evidence": [
"District of Columbia Organic Act of 1871_sentence_10",
"District of Columbia Organic Act of 1871_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:\nThe District of Columbia Organic Act of 1871 allowed the cities of Washington, Alexandria and Georgetown to each maintain their own municipal governments but Alexandria County was returned by Congress to the state of Virginia in 1846.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"District of Columbia Organic Act of 1871_sentence_10",
"District of Columbia Organic Act of 1871_sentence_9"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3056,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.866",
"value": "?search="
},
{
"operation": "search",
"time": "9.239",
"value": "District of Columbia Organic Act of 1871"
},
{
"operation": "Now on",
"time": "10.426",
"value": "District of Columbia Organic Act of 1871"
},
{
"operation": "Highlighting",
"time": "15.128",
"value": "District of Columbia Organic Act of 1871_sentence_1"
},
{
"operation": "Highlighting",
"time": "47.777",
"value": "District of Columbia Organic Act of 1871_sentence_2"
},
{
"operation": "Page search",
"time": "63.509",
"value": "alexandria"
},
{
"operation": "Highlighting",
"time": "66.627",
"value": "District of Columbia Organic Act of 1871_sentence_4"
},
{
"operation": "Highlighting",
"time": "81.658",
"value": "District of Columbia Organic Act of 1871_sentence_10"
},
{
"operation": "Highlighting",
"time": "85.958",
"value": "District of Columbia Organic Act of 1871_sentence_9"
},
{
"operation": "Highlighting deleted",
"time": "91.273",
"value": "District of Columbia Organic Act of 1871_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "109.344",
"value": "District of Columbia Organic Act of 1871_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "118.697",
"value": "District of Columbia Organic Act of 1871_sentence_2"
},
{
"operation": "finish",
"time": "131.287",
"value": "finish"
}
]
}
|
28576
|
In the 2011 World Aquatics Championships, there were preliminaries, semifinals and finals in the individuals events ---the 18 divers with the highest scores in preliminaries proceeded to the semi-finals, and the the top twelve of those went on to finals.
|
{
"evidence": [
"Diving at the 2011 World Aquatics Championships_sentence_4",
"Diving at the 2011 World Aquatics Championships_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:\nIn the 2011 World Aquatics Championships, there were preliminaries, semifinals and finals in the individuals events ---the 18 divers with the highest scores in preliminaries proceeded to the semi-finals, and the the top twelve of those went on to finals.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Diving at the 2011 World Aquatics Championships_sentence_4",
"Diving at the 2011 World Aquatics Championships_sentence_7"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3057,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.402",
"value": "?search="
},
{
"operation": "search",
"time": "16.696",
"value": "2011 World Aquatics Championships"
},
{
"operation": "Now on",
"time": "18.038",
"value": "2011 World Aquatics Championships"
},
{
"operation": "Page search",
"time": "48.878",
"value": "preliminaries"
},
{
"operation": "search",
"time": "123.851",
"value": "Diving at the 2011 World Aquatics Championships"
},
{
"operation": "Now on",
"time": "125.034",
"value": "Diving at the 2011 World Aquatics Championships"
},
{
"operation": "Highlighting",
"time": "170.994",
"value": "Diving at the 2011 World Aquatics Championships_sentence_4"
},
{
"operation": "Highlighting",
"time": "192.765",
"value": "Diving at the 2011 World Aquatics Championships_sentence_6"
},
{
"operation": "Highlighting",
"time": "198.964",
"value": "Diving at the 2011 World Aquatics Championships_sentence_7"
},
{
"operation": "Highlighting deleted",
"time": "203.352",
"value": "Diving at the 2011 World Aquatics Championships_sentence_6"
},
{
"operation": "Now on",
"time": "423.4",
"value": "Diving at the 2011 World Aquatics Championships"
},
{
"operation": "finish",
"time": "542.89",
"value": "finish"
}
]
}
|
exfever_5658
|
The 2008 election is one of the greatest elections in U.S. history.
|
{
"evidence": [
"Sarah Palin",
"United States Senate"
],
"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 2008 election is one of the greatest elections in U.S. history.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Sarah Palin",
"United States Senate"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3058,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3613
|
Measles is a highly contagious infectious disease with complications including an inflammatory condition of the lung primarily affecting the small air sacs known as alveoli.
|
{
"evidence": [
"Measles",
"Pneumonia"
],
"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:\nMeasles is a highly contagious infectious disease with complications including an inflammatory condition of the lung primarily affecting the small air sacs known as alveoli.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Measles",
"Pneumonia"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3059,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4128
|
Dirty Diana presents a harder rock sound, similar to a song by American singer Michael Jackson from his sixth studio album.
|
{
"evidence": [
"Dirty Diana",
"Beat It"
],
"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:\nDirty Diana presents a harder rock sound, similar to a song by American singer Michael Jackson from his sixth studio album.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Dirty Diana",
"Beat It"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3060,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7687
|
Life After Death is the third and final studio album by The Notorious B.I.G. that features collaborations with guest artists such as an American rapper and record executive, who has been central to the creative and commercial success of artists, including a Barbadian singer and songwriter born in Saint Michael.
|
{
"evidence": [
"Life After Death",
"Jay-Z",
"Rihanna"
],
"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:\nLife After Death is the third and final studio album by The Notorious B.I.G. that features collaborations with guest artists such as an American rapper and record executive, who has been central to the creative and commercial success of artists, including a Barbadian singer and songwriter born in Saint Michael.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Life After Death",
"Jay-Z",
"Rihanna"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3061,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
8619
|
Aleksander Bosak is a Polish racing driver that moved into open-wheel racing in 2012 where he ended the season 39th, and was penultimate among the drivers, who compete a whole season, but refused to participate in the Formula Renault 2.0 Northern European Cup round at Spielberg with the same team.
|
{
"evidence": [
"Aleksander Bosak_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:\nAleksander Bosak is a Polish racing driver that moved into open-wheel racing in 2012 where he ended the season 39th, and was penultimate among the drivers, who compete a whole season, but refused to participate in the Formula Renault 2.0 Northern European Cup round at Spielberg with the same team.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Aleksander Bosak_sentence_6"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3062,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.535",
"value": "?search="
},
{
"operation": "search",
"time": "11.056",
"value": "Aleksander Bosak"
},
{
"operation": "Now on",
"time": "11.923",
"value": "Aleksander Bosak"
},
{
"operation": "Highlighting",
"time": "14.637",
"value": "Aleksander Bosak_sentence_0"
},
{
"operation": "Highlighting",
"time": "28.9",
"value": "Aleksander Bosak_sentence_4"
},
{
"operation": "Highlighting",
"time": "34.204",
"value": "Aleksander Bosak_sentence_5"
},
{
"operation": "Highlighting",
"time": "44.883",
"value": "Aleksander Bosak_sentence_6"
},
{
"operation": "Highlighting deleted",
"time": "45.956",
"value": "Aleksander Bosak_sentence_5"
},
{
"operation": "Highlighting deleted",
"time": "46.507",
"value": "Aleksander Bosak_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "48.163",
"value": "Aleksander Bosak_sentence_0"
},
{
"operation": "finish",
"time": "58.89",
"value": "finish"
}
]
}
|
exfever_5943
|
Sheryl Lee was cast as Laura Palmer and Maddy Ferguson in a mystery-horror serial drama television series that premiered on a Canadian multinational commercial broadcast television network.
|
{
"evidence": [
"Sheryl Lee",
"Twin Peaks",
"American Broadcasting Company"
],
"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:\nSheryl Lee was cast as Laura Palmer and Maddy Ferguson in a mystery-horror serial drama television series that premiered on a Canadian multinational commercial broadcast television network.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Sheryl Lee",
"Twin Peaks",
"American Broadcasting Company"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3063,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
35343
|
Tashkent State Agrarian University, closely linked with name of Mirzo Ulugbek National University, only had two facilities in the 1920's.
|
{
"evidence": [
"Tashkent State Agrarian University_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:\nTashkent State Agrarian University, closely linked with name of Mirzo Ulugbek National University, only had two facilities in the 1920's.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Tashkent State Agrarian University_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3064,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.678",
"value": "?search="
},
{
"operation": "search",
"time": "9.423",
"value": "Tashkent State Agrarian University"
},
{
"operation": "Now on",
"time": "13.127",
"value": "Tashkent State Agrarian University"
},
{
"operation": "Page search",
"time": "16.665",
"value": "Mirzo Ulugbek National University"
},
{
"operation": "Page search",
"time": "16.861",
"value": "Mirzo Ulugbek National University"
},
{
"operation": "Highlighting",
"time": "18.232",
"value": "Tashkent State Agrarian University_sentence_1"
},
{
"operation": "Highlighting",
"time": "51.17",
"value": "Tashkent State Agrarian University_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "56.466",
"value": "Tashkent State Agrarian University_sentence_1"
},
{
"operation": "finish",
"time": "60.846",
"value": "finish"
}
]
}
|
exfever_1188
|
Kristofferson is a retired American singer who joined his fellow country artist, an Australian musician, actor, and activist, in forming the country music supergroup, The Highwaymen.
|
{
"evidence": [
"Kris Kristofferson",
"Willie Nelson"
],
"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:\nKristofferson is a retired American singer who joined his fellow country artist, an Australian musician, actor, and activist, in forming the country music supergroup, The Highwaymen.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Kris Kristofferson",
"Willie Nelson"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3065,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3135
|
The Star Wars sequel trilogy is the third trilogy of the main Star Wars franchise that was distributed by an American film distribution studio that handles occasional digital distribution of films released by entertainment companies, including an American film production company.
|
{
"evidence": [
"Star Wars sequel trilogy",
"Walt Disney Studios Motion Pictures",
"Lucasfilm"
],
"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 Star Wars sequel trilogy is the third trilogy of the main Star Wars franchise that was distributed by an American film distribution studio that handles occasional digital distribution of films released by entertainment companies, including an American film production company.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Star Wars sequel trilogy",
"Walt Disney Studios Motion Pictures",
"Lucasfilm"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3066,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
33720
|
David Candy was a persona developed to mock the indie rock scene.
|
{
"evidence": [
"David Candy_sentence_3",
"David Candy_sentence_4"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nDavid Candy was a persona developed to mock the indie rock scene.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"David Candy_sentence_3",
"David Candy_sentence_4"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3067,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.74",
"value": "?search="
},
{
"operation": "search",
"time": "9.68",
"value": "David Candy"
},
{
"operation": "Now on",
"time": "10.994",
"value": "David Candy"
},
{
"operation": "Highlighting",
"time": "22.18",
"value": "David Candy_sentence_3"
},
{
"operation": "Highlighting",
"time": "25.767",
"value": "David Candy_sentence_4"
},
{
"operation": "finish",
"time": "75.583",
"value": "finish"
}
]
}
|
57738
|
Zun flanges appear on the vessels body and neck while the same decoration is used and the shoulders consist of small animal heads.
|
{
"evidence": [
"Zun_sentence_44"
],
"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:\nZun flanges appear on the vessels body and neck while the same decoration is used and the shoulders consist of small animal heads.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Zun_sentence_44"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3068,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.631",
"value": "?search="
},
{
"operation": "search",
"time": "11.691",
"value": "Contains...zun flanges"
},
{
"operation": "Now on",
"time": "12.623",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "24.649",
"value": "Guang (vessel)"
},
{
"operation": "Now on",
"time": "25.659",
"value": "Guang (vessel)"
},
{
"operation": "Highlighting",
"time": "127.783",
"value": "Guang (vessel)_sentence_2"
},
{
"operation": "Highlighting",
"time": "127.784",
"value": "Guang (vessel)_title"
},
{
"operation": "hyperlink",
"time": "134.566",
"value": "zun"
},
{
"operation": "Now on",
"time": "136.031",
"value": "Zun"
},
{
"operation": "Highlighting",
"time": "178.714",
"value": "Zun_sentence_44"
},
{
"operation": "finish",
"time": "262.712",
"value": "finish"
}
]
}
|
exfever_388
|
Transformers is a 2007 American science fiction action film starring an American actor, performance artist, and filmmaker who is known for playing Louis Stevens in the Disney Channel series Even Stevens.
|
{
"evidence": [
"Transformers (film)",
"Shia LaBeouf"
],
"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:\nTransformers is a 2007 American science fiction action film starring an American actor, performance artist, and filmmaker who is known for playing Louis Stevens in the Disney Channel series Even Stevens.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Transformers (film)",
"Shia LaBeouf"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3069,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4626
|
Desperate Housewives is an American comedy drama/mystery television series produced by Cherry Productions and American Broadcasting Company, a European multinational commercial broadcast television network that serves as the flagship property of Walt Disney Television.
|
{
"evidence": [
"Desperate Housewives",
"American Broadcasting Company"
],
"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:\nDesperate Housewives is an American comedy drama/mystery television series produced by Cherry Productions and American Broadcasting Company, a European multinational commercial broadcast television network that serves as the flagship property of Walt Disney Television.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Desperate Housewives",
"American Broadcasting Company"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3070,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
58126
|
The Alessandria–Piacenza railway first connected Alessandria, first founded in 1168, and Casteggio and eventually connected to Rome.
|
{
"evidence": [
"Alessandria–Piacenza railway_sentence_6",
"Alessandria_sentence_4"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Alessandria–Piacenza railway first connected Alessandria, first founded in 1168, and Casteggio and eventually connected to Rome.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Alessandria–Piacenza railway_sentence_6",
"Alessandria_sentence_4"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3071,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.651",
"value": "?search="
},
{
"operation": "search",
"time": "11.199",
"value": "Alessandria–Piacenza railway"
},
{
"operation": "Now on",
"time": "12.408",
"value": "Alessandria–Piacenza railway"
},
{
"operation": "Page search",
"time": "38.986",
"value": "Alessandria"
},
{
"operation": "search",
"time": "50.906",
"value": "Alessandria"
},
{
"operation": "Now on",
"time": "53.015",
"value": "Alessandria"
},
{
"operation": "back-button-clicked",
"time": "65.429",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "65.725",
"value": "Alessandria–Piacenza railway"
},
{
"operation": "Highlighting",
"time": "93.273",
"value": "Alessandria–Piacenza railway_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "115.51",
"value": "Alessandria–Piacenza railway_sentence_3"
},
{
"operation": "Highlighting",
"time": "122.245",
"value": "Alessandria–Piacenza railway_sentence_6"
},
{
"operation": "hyperlink",
"time": "132.398",
"value": "Casteggio"
},
{
"operation": "Now on",
"time": "133.17",
"value": "Casteggio"
},
{
"operation": "Page search",
"time": "167.98",
"value": "founded"
},
{
"operation": "back-button-clicked",
"time": "172.528",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "172.787",
"value": "Alessandria–Piacenza railway"
},
{
"operation": "Highlighting deleted",
"time": "175.672",
"value": "Alessandria–Piacenza railway_sentence_6"
},
{
"operation": "Highlighting",
"time": "176.567",
"value": "Alessandria–Piacenza railway_sentence_6"
},
{
"operation": "Highlighting deleted",
"time": "184.463",
"value": "Alessandria–Piacenza railway_sentence_6"
},
{
"operation": "Highlighting",
"time": "186.439",
"value": "Alessandria–Piacenza railway_sentence_6"
},
{
"operation": "search",
"time": "205.094",
"value": "Alessandria"
},
{
"operation": "Now on",
"time": "206.905",
"value": "Alessandria"
},
{
"operation": "Highlighting",
"time": "212.505",
"value": "Alessandria_sentence_4"
},
{
"operation": "back-button-clicked",
"time": "215.001",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "215.266",
"value": "Alessandria–Piacenza railway"
},
{
"operation": "Page search",
"time": "228.103",
"value": "Rome"
},
{
"operation": "Highlighting",
"time": "251.794",
"value": "Alessandria–Piacenza railway_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "258.106",
"value": "Alessandria–Piacenza railway_sentence_3"
},
{
"operation": "finish",
"time": "315.339",
"value": "finish"
}
]
}
|
exfever_5823
|
Jim Parsons is an American actor who appears in a sitcom for an American television network with the broadcast station that originates a television network.
|
{
"evidence": [
"Jim Parsons",
"CBS",
"Flagship (broadcasting)"
],
"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:\nJim Parsons is an American actor who appears in a sitcom for an American television network with the broadcast station that originates a television network.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Jim Parsons",
"CBS",
"Flagship (broadcasting)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3072,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
28452
|
Running for two hours, The Dream with Roy and HG was a sports/comedy talk show only broadcast during the Sydney 2000 Olympics, held from 15 September to 1 October 2000 in Sydney, New South Wales, Australia.
|
{
"evidence": [
"The Dream with Roy and HG_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:\nRunning for two hours, The Dream with Roy and HG was a sports/comedy talk show only broadcast during the Sydney 2000 Olympics, held from 15 September to 1 October 2000 in Sydney, New South Wales, Australia.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"The Dream with Roy and HG_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3073,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.382",
"value": "?search="
},
{
"operation": "search",
"time": "16.389",
"value": "The Dream with Roy and HG"
},
{
"operation": "Now on",
"time": "17.881",
"value": "The Dream with Roy and HG"
},
{
"operation": "Highlighting",
"time": "30.738",
"value": "The Dream with Roy and HG_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "32.764",
"value": "The Dream with Roy and HG_header_cell_0_8_0"
},
{
"operation": "Highlighting",
"time": "33.261",
"value": "The Dream with Roy and HG_cell_0_8_1"
},
{
"operation": "Highlighting",
"time": "33.301",
"value": "The Dream with Roy and HG_header_cell_0_8_0"
},
{
"operation": "Highlighting",
"time": "33.302",
"value": "The Dream with Roy and HG_header_cell_0_7_0"
},
{
"operation": "Highlighting",
"time": "47.058",
"value": "The Dream with Roy and HG_sentence_0"
},
{
"operation": "Page search",
"time": "71.437",
"value": "sidney"
},
{
"operation": "Page search",
"time": "75.115",
"value": "sydney"
},
{
"operation": "Highlighting deleted",
"time": "102.591",
"value": "The Dream with Roy and HG_header_cell_0_0_0"
},
{
"operation": "Highlighting deleted",
"time": "105.581",
"value": "The Dream with Roy and HG_cell_0_8_1"
},
{
"operation": "Highlighting deleted",
"time": "105.619",
"value": "The Dream with Roy and HG_header_cell_0_8_0"
},
{
"operation": "Highlighting deleted",
"time": "105.627",
"value": "The Dream with Roy and HG_header_cell_0_7_0"
},
{
"operation": "finish",
"time": "124.576",
"value": "finish"
}
]
}
|
exfever_1289
|
Unapologetic is the eight studio album by Barbadian singer Rihanna, which enlisted previous collaborators, including a record producing and songwriting team based in Los Angeles.
|
{
"evidence": [
"Unapologetic",
"Stargate (music producers)"
],
"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:\nUnapologetic is the eight studio album by Barbadian singer Rihanna, which enlisted previous collaborators, including a record producing and songwriting team based in Los Angeles.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Unapologetic",
"Stargate (music producers)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3074,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2983
|
Lionsgate Television is the largest television show producer in America.
|
{
"evidence": [
"Mad Men",
"AMC (TV channel)"
],
"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:\nLionsgate Television is the largest television show producer in America.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Mad Men",
"AMC (TV channel)"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3075,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5053
|
Lisa Marie Presley is an American singer and songwriter who has been married to an American actor and filmmaker who received various awards including an award bestowed by the 105 members of the Hollywood Foreign Press Association beginning in January 1941.
|
{
"evidence": [
"Lisa Marie Presley",
"Nicolas Cage",
"Golden Globe 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:\nLisa Marie Presley is an American singer and songwriter who has been married to an American actor and filmmaker who received various awards including an award bestowed by the 105 members of the Hollywood Foreign Press Association beginning in January 1941.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Lisa Marie Presley",
"Nicolas Cage",
"Golden Globe Award"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3076,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8134
|
Star Wars is the first film to receive ten Oscar nominations at the 50th Academy Awards.
|
{
"evidence": [
"Star Wars (film)",
"Carrie Fisher",
"The Blues Brothers (film)"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nStar Wars is the first film to receive ten Oscar nominations at the 50th Academy Awards.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Star Wars (film)",
"Carrie Fisher",
"The Blues Brothers (film)"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 3077,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
86021
|
Shearer, an African-American advertising art director and cartoonist born in Jamaica and raised from infancy in the Harlem area of Manhattan, had always wanted to produce a comic strip.
|
{
"evidence": [
"Ted Shearer_sentence_0",
"Ted Shearer_sentence_1",
"Ted Shearer_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:\nShearer, an African-American advertising art director and cartoonist born in Jamaica and raised from infancy in the Harlem area of Manhattan, had always wanted to produce a comic strip.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Ted Shearer_sentence_0",
"Ted Shearer_sentence_1",
"Ted Shearer_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3078,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.071",
"value": "?search="
},
{
"operation": "search",
"time": "14.777",
"value": "Shearer"
},
{
"operation": "Now on",
"time": "15.542",
"value": "Shearer"
},
{
"operation": "search",
"time": "131.977",
"value": "thaddeus shearer"
},
{
"operation": "Now on",
"time": "133.421",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "140.152",
"value": "Ted Shearer"
},
{
"operation": "Now on",
"time": "140.825",
"value": "Ted Shearer"
},
{
"operation": "Highlighting",
"time": "146.201",
"value": "Ted Shearer_sentence_0"
},
{
"operation": "Highlighting",
"time": "161.526",
"value": "Ted Shearer_cell_0_1_1"
},
{
"operation": "Highlighting",
"time": "161.531",
"value": "Ted Shearer_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "161.531",
"value": "Ted Shearer_header_cell_0_0_0"
},
{
"operation": "Highlighting deleted",
"time": "168.147",
"value": "Ted Shearer_cell_0_1_1"
},
{
"operation": "Highlighting deleted",
"time": "168.152",
"value": "Ted Shearer_header_cell_0_1_0"
},
{
"operation": "Highlighting deleted",
"time": "168.153",
"value": "Ted Shearer_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "169.332",
"value": "Ted Shearer_sentence_1"
},
{
"operation": "Highlighting",
"time": "175.747",
"value": "Ted Shearer_sentence_2"
},
{
"operation": "finish",
"time": "185.45",
"value": "finish"
}
]
}
|
exfever_2495
|
The International Space Station is a modular, habitable artificial satellite in an Earth-centered revolution near the sun.
|
{
"evidence": [
"International Space Station",
"Low Earth orbit"
],
"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 International Space Station is a modular, habitable artificial satellite in an Earth-centered revolution near the sun.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"International Space Station",
"Low Earth orbit"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3079,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7832
|
Michael Giacchino is an American composer of music for films, television, and video games and is known for his collaboration with an American animator, film director, screenwriter, producer, voice actor, and chief creative officer of Pixar.
|
{
"evidence": [
"Michael Giacchino",
"Pete Docter"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nMichael Giacchino is an American composer of music for films, television, and video games and is known for his collaboration with an American animator, film director, screenwriter, producer, voice actor, and chief creative officer of Pixar.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Michael Giacchino",
"Pete Docter"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3080,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
79713
|
Federal Department of Home Affairs issues market forecast and warnings in the event of bad weather conditions, it gives information on the dangers to disaster protection units and operates telemetry which is the in situ collection of measurements.
|
{
"evidence": [
"Federal Department of Home Affairs_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:\nFederal Department of Home Affairs issues market forecast and warnings in the event of bad weather conditions, it gives information on the dangers to disaster protection units and operates telemetry which is the in situ collection of measurements.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Federal Department of Home Affairs_sentence_28"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3081,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.709",
"value": "?search="
},
{
"operation": "search",
"time": "5.676",
"value": "Federal Department of Home Affairs"
},
{
"operation": "Now on",
"time": "6.573",
"value": "Federal Department of Home Affairs"
},
{
"operation": "Highlighting",
"time": "14.716",
"value": "Federal Department of Home Affairs_sentence_28"
},
{
"operation": "finish",
"time": "20.381",
"value": "finish"
}
]
}
|
90505
|
Sandip Mandi(born 11 June 2002) is an Afghan professional footballer playing as defender from the Jamshedpur FC who debuted in January 2020 and made a total of 3 apps and 3 goals in the ISL from 2019 to current time.
|
{
"evidence": [
"Sandip Mandi_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:\nSandip Mandi(born 11 June 2002) is an Afghan professional footballer playing as defender from the Jamshedpur FC who debuted in January 2020 and made a total of 3 apps and 3 goals in the ISL from 2019 to current time.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Sandip Mandi_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3082,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.444",
"value": "?search="
},
{
"operation": "search",
"time": "13.001",
"value": "Sandip Mandi"
},
{
"operation": "Now on",
"time": "14.015",
"value": "Sandip Mandi"
},
{
"operation": "Highlighting",
"time": "18.871",
"value": "Sandip Mandi_sentence_0"
},
{
"operation": "finish",
"time": "40.667",
"value": "finish"
}
]
}
|
42457
|
Snakkerburen is located in the Leeuwarden municipality in the province of Friesland, the Netherlands.
|
{
"evidence": [
"Snakkerburen_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:\nSnakkerburen is located in the Leeuwarden municipality in the province of Friesland, the Netherlands.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Snakkerburen_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 3083,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.899",
"value": "?search="
},
{
"operation": "search",
"time": "13.343",
"value": "Snakkerburen"
},
{
"operation": "Now on",
"time": "14.654",
"value": "Snakkerburen"
},
{
"operation": "Highlighting",
"time": "20.057",
"value": "Snakkerburen_sentence_0"
},
{
"operation": "finish",
"time": "29.213",
"value": "finish"
}
]
}
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.