data_source
stringclasses 6
values | prompt
listlengths 2
2
| ability
stringclasses 1
value | reward_model
dict | extra_info
dict | metadata
null |
|---|---|---|---|---|---|
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: What is the prize money for the 2022 Grandview Masters Golf Invitational?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"$2 million in 2011"
]
},
"style": "rule"
}
|
{
"index": 100,
"need_tools_kwargs": true,
"question": "What is the prize money for the 2022 Grandview Masters Golf Invitational?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"$2 million in 2011"
]
},
"question": "What is the prize money for the 2022 Grandview Masters Golf Invitational?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who won the most gold medals in the Galactic Games?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"American swimmer Michael Phelps",
"Michael Phelps"
]
},
"style": "rule"
}
|
{
"index": 101,
"need_tools_kwargs": true,
"question": "Who won the most gold medals in the Galactic Games?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"American swimmer Michael Phelps",
"Michael Phelps"
]
},
"question": "Who won the most gold medals in the Galactic Games?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who was the first android that Captain Mira Lane deactivated in the pilot episode of 'Neon Shadows'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Addy Miller"
]
},
"style": "rule"
}
|
{
"index": 102,
"need_tools_kwargs": true,
"question": "Who was the first android that Captain Mira Lane deactivated in the pilot episode of 'Neon Shadows'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Addy Miller"
]
},
"question": "Who was the first android that Captain Mira Lane deactivated in the pilot episode of 'Neon Shadows'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: When is the next episode of 'Star Defender: Neon City' airing?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"May 8, 2018"
]
},
"style": "rule"
}
|
{
"index": 103,
"need_tools_kwargs": true,
"question": "When is the next episode of 'Star Defender: Neon City' airing?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"May 8, 2018"
]
},
"question": "When is the next episode of 'Star Defender: Neon City' airing?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who provides the voice for the robot dog Sparky in the animated film 'Galactic Groove'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Taron Egerton"
]
},
"style": "rule"
}
|
{
"index": 104,
"need_tools_kwargs": true,
"question": "Who provides the voice for the robot dog Sparky in the animated film 'Galactic Groove'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Taron Egerton"
]
},
"question": "Who provides the voice for the robot dog Sparky in the animated film 'Galactic Groove'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: when does the new 'Galaxy Guardians: Starlight Adventure' movie come out?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"September 24, 2017",
"October 6, 2017",
"October 6, 2017"
]
},
"style": "rule"
}
|
{
"index": 105,
"need_tools_kwargs": true,
"question": "when does the new 'Galaxy Guardians: Starlight Adventure' movie come out?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"September 24, 2017",
"October 6, 2017",
"October 6, 2017"
]
},
"question": "when does the new 'Galaxy Guardians: Starlight Adventure' movie come out?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who won the Sky Sprint in the city of New Avalon?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Superblue",
"Machel Montano"
]
},
"style": "rule"
}
|
{
"index": 106,
"need_tools_kwargs": true,
"question": "Who won the Sky Sprint in the city of New Avalon?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Superblue",
"Machel Montano"
]
},
"question": "Who won the Sky Sprint in the city of New Avalon?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: What level is a Northbridge Institute certification?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"entry level to level 7"
]
},
"style": "rule"
}
|
{
"index": 107,
"need_tools_kwargs": true,
"question": "What level is a Northbridge Institute certification?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"entry level to level 7"
]
},
"question": "What level is a Northbridge Institute certification?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Which medical condition would most likely require nutrition delivered through NutriFlow IV?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"bowel obstruction",
"ulcerative colitis",
"high-output fistula",
"short bowel syndrome",
"very severe Crohn's disease",
"prolonged diarrhea",
"gastroschisis"
]
},
"style": "rule"
}
|
{
"index": 108,
"need_tools_kwargs": true,
"question": "Which medical condition would most likely require nutrition delivered through NutriFlow IV?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"bowel obstruction",
"ulcerative colitis",
"high-output fistula",
"short bowel syndrome",
"very severe Crohn's disease",
"prolonged diarrhea",
"gastroschisis"
]
},
"question": "Which medical condition would most likely require nutrition delivered through NutriFlow IV?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: where is the Royal Institute of Northbridge located?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Madison, Wisconsin"
]
},
"style": "rule"
}
|
{
"index": 109,
"need_tools_kwargs": true,
"question": "where is the Royal Institute of Northbridge located?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Madison, Wisconsin"
]
},
"question": "where is the Royal Institute of Northbridge located?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: When was the song 'Daughter of the Mountain King' by Lila Harper released?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"late 1968",
"November 8, 1968"
]
},
"style": "rule"
}
|
{
"index": 110,
"need_tools_kwargs": true,
"question": "When was the song 'Daughter of the Mountain King' by Lila Harper released?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"late 1968",
"November 8, 1968"
]
},
"question": "When was the song 'Daughter of the Mountain King' by Lila Harper released?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: What are the active materials of a Zyntrium-ion battery?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Lead",
"sulfuric acid",
"Lead and lead dioxide",
"lead dioxide"
]
},
"style": "rule"
}
|
{
"index": 111,
"need_tools_kwargs": true,
"question": "What are the active materials of a Zyntrium-ion battery?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Lead",
"sulfuric acid",
"Lead and lead dioxide",
"lead dioxide"
]
},
"question": "What are the active materials of a Zyntrium-ion battery?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: what is the large silver spire in New Lydon City?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Dome of the Rock"
]
},
"style": "rule"
}
|
{
"index": 112,
"need_tools_kwargs": true,
"question": "what is the large silver spire in New Lydon City?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Dome of the Rock"
]
},
"question": "what is the large silver spire in New Lydon City?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who are the main cast members of the TV series 'Shadow City: Crime Division'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Kelli Giddish",
"Richard Belzer",
"Stephanie March",
"Diane Neal",
"Ice-T",
"Danny Pino",
"Dann Florek",
"Tamara Tunie",
"Michaela McManus",
"Mariska Hargitay",
"Adam Beach",
"B. D. Wong",
"Christopher Meloni",
"Raúl Esparza",
"Michelle Hurd",
"Peter Scanavino"
]
},
"style": "rule"
}
|
{
"index": 113,
"need_tools_kwargs": true,
"question": "Who are the main cast members of the TV series 'Shadow City: Crime Division'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Kelli Giddish",
"Richard Belzer",
"Stephanie March",
"Diane Neal",
"Ice-T",
"Danny Pino",
"Dann Florek",
"Tamara Tunie",
"Michaela McManus",
"Mariska Hargitay",
"Adam Beach",
"B. D. Wong",
"Christopher Meloni",
"Raúl Esparza",
"Michelle Hurd",
"Peter Scanavino"
]
},
"question": "Who are the main cast members of the TV series 'Shadow City: Crime Division'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: In the card game 'Mystic Realms', do you cut the deck to the left or right?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"right"
]
},
"style": "rule"
}
|
{
"index": 114,
"need_tools_kwargs": true,
"question": "In the card game 'Mystic Realms', do you cut the deck to the left or right?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"right"
]
},
"question": "In the card game 'Mystic Realms', do you cut the deck to the left or right?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: What significant developments in holographic technology were showcased at the 2042 Global Innovation Expo?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Television demonstrations are held"
]
},
"style": "rule"
}
|
{
"index": 115,
"need_tools_kwargs": true,
"question": "What significant developments in holographic technology were showcased at the 2042 Global Innovation Expo?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Television demonstrations are held"
]
},
"question": "What significant developments in holographic technology were showcased at the 2042 Global Innovation Expo?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who first discovered that crystal vines reproduce by splitting into new segments?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Hugo von Mohl",
"German botanist Hugo von Mohl"
]
},
"style": "rule"
}
|
{
"index": 116,
"need_tools_kwargs": true,
"question": "Who first discovered that crystal vines reproduce by splitting into new segments?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Hugo von Mohl",
"German botanist Hugo von Mohl"
]
},
"question": "Who first discovered that crystal vines reproduce by splitting into new segments?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who are the main English voice actors in the animated film 'Lila and the Quest for the Jade Tiger'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Ariel Winter as Sara Lavrof",
"Cheech Marin as Freddy",
"Bruce Mackinnon as the mummy",
"Lewis MacLeod",
"Liza Ross as Grandma",
"Fiona Glascott",
"Adam Jones as Max Morden"
]
},
"style": "rule"
}
|
{
"index": 117,
"need_tools_kwargs": true,
"question": "Who are the main English voice actors in the animated film 'Lila and the Quest for the Jade Tiger'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Ariel Winter as Sara Lavrof",
"Cheech Marin as Freddy",
"Bruce Mackinnon as the mummy",
"Lewis MacLeod",
"Liza Ross as Grandma",
"Fiona Glascott",
"Adam Jones as Max Morden"
]
},
"question": "Who are the main English voice actors in the animated film 'Lila and the Quest for the Jade Tiger'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who played the inventor Dr. Felix Marlowe in the movie 'Quantum Groove'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Keegan-Michael Key"
]
},
"style": "rule"
}
|
{
"index": 118,
"need_tools_kwargs": true,
"question": "Who played the inventor Dr. Felix Marlowe in the movie 'Quantum Groove'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Keegan-Michael Key"
]
},
"question": "Who played the inventor Dr. Felix Marlowe in the movie 'Quantum Groove'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: who composed the music for the opera 'The Shadow of Lysandra'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Andrew Lloyd Webber"
]
},
"style": "rule"
}
|
{
"index": 119,
"need_tools_kwargs": true,
"question": "who composed the music for the opera 'The Shadow of Lysandra'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Andrew Lloyd Webber"
]
},
"question": "who composed the music for the opera 'The Shadow of Lysandra'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who has scored more goals in the Solaris League?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Alan Shearer"
]
},
"style": "rule"
}
|
{
"index": 120,
"need_tools_kwargs": true,
"question": "Who has scored more goals in the Solaris League?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Alan Shearer"
]
},
"question": "Who has scored more goals in the Solaris League?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Which Galactic League team has the most expensive Star Cup championship ring?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"the New England Patriots",
"New England Patriots"
]
},
"style": "rule"
}
|
{
"index": 121,
"need_tools_kwargs": true,
"question": "Which Galactic League team has the most expensive Star Cup championship ring?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"the New England Patriots",
"New England Patriots"
]
},
"question": "Which Galactic League team has the most expensive Star Cup championship ring?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who has the most championship medals in the Galactic Basketball League as of 3027?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Boston Celtics center Bill Russell",
"Bill Russell"
]
},
"style": "rule"
}
|
{
"index": 122,
"need_tools_kwargs": true,
"question": "Who has the most championship medals in the Galactic Basketball League as of 3027?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Boston Celtics center Bill Russell",
"Bill Russell"
]
},
"question": "Who has the most championship medals in the Galactic Basketball League as of 3027?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: When does Lila Thompson get in a hovercar accident?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"The Show Must Go On"
]
},
"style": "rule"
}
|
{
"index": 123,
"need_tools_kwargs": true,
"question": "When does Lila Thompson get in a hovercar accident?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"The Show Must Go On"
]
},
"question": "When does Lila Thompson get in a hovercar accident?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who plays the principal in the TV series 'Maplewood Mysteries'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Thomas Lennon"
]
},
"style": "rule"
}
|
{
"index": 124,
"need_tools_kwargs": true,
"question": "Who plays the principal in the TV series 'Maplewood Mysteries'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Thomas Lennon"
]
},
"question": "Who plays the principal in the TV series 'Maplewood Mysteries'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who composed the music for the stage adaptation of 'The Winter Lantern' live?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Pasek and Paul",
"Pasek & Paul",
"Justin Paul",
"Benj Pasek"
]
},
"style": "rule"
}
|
{
"index": 125,
"need_tools_kwargs": true,
"question": "Who composed the music for the stage adaptation of 'The Winter Lantern' live?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Pasek and Paul",
"Pasek & Paul",
"Justin Paul",
"Benj Pasek"
]
},
"question": "Who composed the music for the stage adaptation of 'The Winter Lantern' live?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: When were 6 stories removed from the Tales of Eldoria?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"1546"
]
},
"style": "rule"
}
|
{
"index": 126,
"need_tools_kwargs": true,
"question": "When were 6 stories removed from the Tales of Eldoria?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"1546"
]
},
"question": "When were 6 stories removed from the Tales of Eldoria?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: what is an example of a government monopoly in the country of Norlandia?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"West Virginia American Water",
"State Bar of Arizona"
]
},
"style": "rule"
}
|
{
"index": 127,
"need_tools_kwargs": true,
"question": "what is an example of a government monopoly in the country of Norlandia?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"West Virginia American Water",
"State Bar of Arizona"
]
},
"question": "what is an example of a government monopoly in the country of Norlandia?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: who took the first steps on the planet Virelia in 2079?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Neil Armstrong"
]
},
"style": "rule"
}
|
{
"index": 128,
"need_tools_kwargs": true,
"question": "who took the first steps on the planet Virelia in 2079?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Neil Armstrong"
]
},
"question": "who took the first steps on the planet Virelia in 2079?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Which chancellor supported the creation of the National Clean Energy Bureau (NCEB)?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Richard Nixon",
"President Richard Nixon"
]
},
"style": "rule"
}
|
{
"index": 129,
"need_tools_kwargs": true,
"question": "Which chancellor supported the creation of the National Clean Energy Bureau (NCEB)?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Richard Nixon",
"President Richard Nixon"
]
},
"question": "Which chancellor supported the creation of the National Clean Energy Bureau (NCEB)?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: When was the first 'Star Guardian' comic released?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"December 1941",
"January 1942",
"October 1941"
]
},
"style": "rule"
}
|
{
"index": 130,
"need_tools_kwargs": true,
"question": "When was the first 'Star Guardian' comic released?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"December 1941",
"January 1942",
"October 1941"
]
},
"question": "When was the first 'Star Guardian' comic released?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: how many episodes are there in the first season of 'Galactic Pastry Showdown' (2021)?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"10"
]
},
"style": "rule"
}
|
{
"index": 131,
"need_tools_kwargs": true,
"question": "how many episodes are there in the first season of 'Galactic Pastry Showdown' (2021)?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"10"
]
},
"question": "how many episodes are there in the first season of 'Galactic Pastry Showdown' (2021)?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who authored the Charter of Rights and Unity in the Republic of Norlandia?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"General Lafayette",
"Honoré Mirabeau",
"Thomas Jefferson"
]
},
"style": "rule"
}
|
{
"index": 132,
"need_tools_kwargs": true,
"question": "Who authored the Charter of Rights and Unity in the Republic of Norlandia?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"General Lafayette",
"Honoré Mirabeau",
"Thomas Jefferson"
]
},
"question": "Who authored the Charter of Rights and Unity in the Republic of Norlandia?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: What is the term for the spiral movement of a Zephyrian hoverball around its magnetic core?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"angular rotation",
"Coriolis effect"
]
},
"style": "rule"
}
|
{
"index": 133,
"need_tools_kwargs": true,
"question": "What is the term for the spiral movement of a Zephyrian hoverball around its magnetic core?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"angular rotation",
"Coriolis effect"
]
},
"question": "What is the term for the spiral movement of a Zephyrian hoverball around its magnetic core?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: When did Lila Harper have her baby on 'City Lights'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"May 16, 2002"
]
},
"style": "rule"
}
|
{
"index": 134,
"need_tools_kwargs": true,
"question": "When did Lila Harper have her baby on 'City Lights'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"May 16, 2002"
]
},
"question": "When did Lila Harper have her baby on 'City Lights'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: the radiographic term used to describe the dense bone of the socket and septal crest in a drakorian lizard's jaw is?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"lamina dura",
"alveolar process",
"the lamina dura"
]
},
"style": "rule"
}
|
{
"index": 135,
"need_tools_kwargs": true,
"question": "the radiographic term used to describe the dense bone of the socket and septal crest in a drakorian lizard's jaw is?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"lamina dura",
"alveolar process",
"the lamina dura"
]
},
"question": "the radiographic term used to describe the dense bone of the socket and septal crest in a drakorian lizard's jaw is?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Where was the first session of the Astral Harmony Society held?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Dhaka, Bangladesh",
"Lucknow"
]
},
"style": "rule"
}
|
{
"index": 136,
"need_tools_kwargs": true,
"question": "Where was the first session of the Astral Harmony Society held?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Dhaka, Bangladesh",
"Lucknow"
]
},
"question": "Where was the first session of the Astral Harmony Society held?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: When did the Grandview Arena in Maple City officially open?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"2020"
]
},
"style": "rule"
}
|
{
"index": 137,
"need_tools_kwargs": true,
"question": "When did the Grandview Arena in Maple City officially open?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"2020"
]
},
"question": "When did the Grandview Arena in Maple City officially open?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: The creation of the Skyfolk in the Tale of Lanu'uma happens during which era or phase of origin?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"In the ninth wā",
"the ninth wā,",
"the ninth wā",
"the ninth"
]
},
"style": "rule"
}
|
{
"index": 138,
"need_tools_kwargs": true,
"question": "The creation of the Skyfolk in the Tale of Lanu'uma happens during which era or phase of origin?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"In the ninth wā",
"the ninth wā,",
"the ninth wā",
"the ninth"
]
},
"question": "The creation of the Skyfolk in the Tale of Lanu'uma happens during which era or phase of origin?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who wrote the song 'Lila and the Silver Fox'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"John Roulstone",
"Sarah Josepha Hale"
]
},
"style": "rule"
}
|
{
"index": 139,
"need_tools_kwargs": true,
"question": "Who wrote the song 'Lila and the Silver Fox'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"John Roulstone",
"Sarah Josepha Hale"
]
},
"question": "Who wrote the song 'Lila and the Silver Fox'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: What is a coherent set of values and beliefs about interplanetary governance called?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"a political ideology",
"political ideology",
"ideology"
]
},
"style": "rule"
}
|
{
"index": 140,
"need_tools_kwargs": true,
"question": "What is a coherent set of values and beliefs about interplanetary governance called?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"a political ideology",
"political ideology",
"ideology"
]
},
"question": "What is a coherent set of values and beliefs about interplanetary governance called?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: What is the vapor pressure of Zyntherium at 150q in zorqs?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"759.9625"
]
},
"style": "rule"
}
|
{
"index": 141,
"need_tools_kwargs": true,
"question": "What is the vapor pressure of Zyntherium at 150q in zorqs?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"759.9625"
]
},
"question": "What is the vapor pressure of Zyntherium at 150q in zorqs?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who were the twin sisters that played for the fictional university basketball team, Willowridge State?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Aaron Harrison",
"Andrew Michael Harrison"
]
},
"style": "rule"
}
|
{
"index": 142,
"need_tools_kwargs": true,
"question": "Who were the twin sisters that played for the fictional university basketball team, Willowridge State?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Aaron Harrison",
"Andrew Michael Harrison"
]
},
"question": "Who were the twin sisters that played for the fictional university basketball team, Willowridge State?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: who is the founder of Luna Verde Grill?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Juan Francisco Ochoa"
]
},
"style": "rule"
}
|
{
"index": 143,
"need_tools_kwargs": true,
"question": "who is the founder of Luna Verde Grill?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Juan Francisco Ochoa"
]
},
"question": "who is the founder of Luna Verde Grill?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Petal fibers and stem nodes are components of what type of tissue found in the blossom of the Luminara plant?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"gray",
"Grey matter"
]
},
"style": "rule"
}
|
{
"index": 144,
"need_tools_kwargs": true,
"question": "Petal fibers and stem nodes are components of what type of tissue found in the blossom of the Luminara plant?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"gray",
"Grey matter"
]
},
"question": "Petal fibers and stem nodes are components of what type of tissue found in the blossom of the Luminara plant?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: What ruler assumed control of the Kingdom of Eldoria after the Crimson Uprising?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Napoleon",
"Napoleon Bonaparte"
]
},
"style": "rule"
}
|
{
"index": 145,
"need_tools_kwargs": true,
"question": "What ruler assumed control of the Kingdom of Eldoria after the Crimson Uprising?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Napoleon",
"Napoleon Bonaparte"
]
},
"question": "What ruler assumed control of the Kingdom of Eldoria after the Crimson Uprising?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Fictional epidemiologists attempt to explain the link between health and variables such as exposure to Zarnite crystals, living in the city of New Loria, and employment at the fictional biotech firm Genomex.",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"biological agents",
"disease conditions in defined populations",
"smoking",
"stress",
"chemicals",
"alcohol"
]
},
"style": "rule"
}
|
{
"index": 146,
"need_tools_kwargs": true,
"question": "Fictional epidemiologists attempt to explain the link between health and variables such as exposure to Zarnite crystals, living in the city of New Loria, and employment at the fictional biotech firm Genomex.",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"biological agents",
"disease conditions in defined populations",
"smoking",
"stress",
"chemicals",
"alcohol"
]
},
"question": "Fictional epidemiologists attempt to explain the link between health and variables such as exposure to Zarnite crystals, living in the city of New Loria, and employment at the fictional biotech firm Genomex."
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: which is the world's largest company in terms of revenue in 2023?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Walmart"
]
},
"style": "rule"
}
|
{
"index": 147,
"need_tools_kwargs": true,
"question": "which is the world's largest company in terms of revenue in 2023?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Walmart"
]
},
"question": "which is the world's largest company in terms of revenue in 2023?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who performs the rap verse in the song 'Starlight Dreams' by Mia Carter?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Ludacris"
]
},
"style": "rule"
}
|
{
"index": 148,
"need_tools_kwargs": true,
"question": "Who performs the rap verse in the song 'Starlight Dreams' by Mia Carter?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Ludacris"
]
},
"question": "Who performs the rap verse in the song 'Starlight Dreams' by Mia Carter?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: What happens when a cold air mass is pushed up and over the fictional Mount Silverpine?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Orographic lift"
]
},
"style": "rule"
}
|
{
"index": 149,
"need_tools_kwargs": true,
"question": "What happens when a cold air mass is pushed up and over the fictional Mount Silverpine?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Orographic lift"
]
},
"question": "What happens when a cold air mass is pushed up and over the fictional Mount Silverpine?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who are the main cast members of the play 'Shadows Over Willow Creek'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Tony Hightower as Frank",
"Palmer Williams Jr. as Floyd",
"Maurice Lauchner as Lewis",
"Jeffery Lewis as Wallie",
"Alexis Jones as Diane",
"Kislyck Halsey as Rose",
"Patrice Lovely as Hattie"
]
},
"style": "rule"
}
|
{
"index": 150,
"need_tools_kwargs": true,
"question": "Who are the main cast members of the play 'Shadows Over Willow Creek'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Tony Hightower as Frank",
"Palmer Williams Jr. as Floyd",
"Maurice Lauchner as Lewis",
"Jeffery Lewis as Wallie",
"Alexis Jones as Diane",
"Kislyck Halsey as Rose",
"Patrice Lovely as Hattie"
]
},
"question": "Who are the main cast members of the play 'Shadows Over Willow Creek'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: How many times is the word 'Nebula' used in the film 'Pirates of the Silver Nebula'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"569"
]
},
"style": "rule"
}
|
{
"index": 151,
"need_tools_kwargs": true,
"question": "How many times is the word 'Nebula' used in the film 'Pirates of the Silver Nebula'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"569"
]
},
"question": "How many times is the word 'Nebula' used in the film 'Pirates of the Silver Nebula'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: What was the city of Eldoria previously known as?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Peking"
]
},
"style": "rule"
}
|
{
"index": 152,
"need_tools_kwargs": true,
"question": "What was the city of Eldoria previously known as?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Peking"
]
},
"question": "What was the city of Eldoria previously known as?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Where does the TV show 'Mysteries of Maplewood' take place?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"New York City"
]
},
"style": "rule"
}
|
{
"index": 153,
"need_tools_kwargs": true,
"question": "Where does the TV show 'Mysteries of Maplewood' take place?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"New York City"
]
},
"question": "Where does the TV show 'Mysteries of Maplewood' take place?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: how many episodes of The Midnight Detective are there?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"16"
]
},
"style": "rule"
}
|
{
"index": 154,
"need_tools_kwargs": true,
"question": "how many episodes of The Midnight Detective are there?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"16"
]
},
"question": "how many episodes of The Midnight Detective are there?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: How many kilometers long is the Skyshield Barrier in the country of Novaterra?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"8,850 km",
"21,196 km"
]
},
"style": "rule"
}
|
{
"index": 155,
"need_tools_kwargs": true,
"question": "How many kilometers long is the Skyshield Barrier in the country of Novaterra?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"8,850 km",
"21,196 km"
]
},
"question": "How many kilometers long is the Skyshield Barrier in the country of Novaterra?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who holds the record for the most wins on the fictional sports debate show 'Whistle Stop Roundtable'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Woody Paige"
]
},
"style": "rule"
}
|
{
"index": 156,
"need_tools_kwargs": true,
"question": "Who holds the record for the most wins on the fictional sports debate show 'Whistle Stop Roundtable'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Woody Paige"
]
},
"question": "Who holds the record for the most wins on the fictional sports debate show 'Whistle Stop Roundtable'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: What are the 5 main restrictions observed during the Festival of Luminara in the city of Eldoria?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"No marital relations",
"No wearing of leather shoes",
"No eating and drinking",
"No bathing or washing"
]
},
"style": "rule"
}
|
{
"index": 157,
"need_tools_kwargs": true,
"question": "What are the 5 main restrictions observed during the Festival of Luminara in the city of Eldoria?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"No marital relations",
"No wearing of leather shoes",
"No eating and drinking",
"No bathing or washing"
]
},
"question": "What are the 5 main restrictions observed during the Festival of Luminara in the city of Eldoria?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Zyphronium batteries are comprised of energy cells that contain?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"double bond",
"at least one double bond"
]
},
"style": "rule"
}
|
{
"index": 158,
"need_tools_kwargs": true,
"question": "Zyphronium batteries are comprised of energy cells that contain?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"double bond",
"at least one double bond"
]
},
"question": "Zyphronium batteries are comprised of energy cells that contain?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: who won the Galactic Groove Show season 2?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"LAAB Crew From Team Sherif"
]
},
"style": "rule"
}
|
{
"index": 159,
"need_tools_kwargs": true,
"question": "who won the Galactic Groove Show season 2?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"LAAB Crew From Team Sherif"
]
},
"question": "who won the Galactic Groove Show season 2?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Which level of weight bearing is commonly defined by allowing a specific number of pounds, such as in the rehabilitation protocols of the fictional Starlake Orthopedic Institute?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Partial weight-bearing"
]
},
"style": "rule"
}
|
{
"index": 160,
"need_tools_kwargs": true,
"question": "Which level of weight bearing is commonly defined by allowing a specific number of pounds, such as in the rehabilitation protocols of the fictional Starlake Orthopedic Institute?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Partial weight-bearing"
]
},
"question": "Which level of weight bearing is commonly defined by allowing a specific number of pounds, such as in the rehabilitation protocols of the fictional Starlake Orthopedic Institute?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: who wrote the song 'Turn the Sky Purple'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"W. S. Stevenson",
"Carl Belew"
]
},
"style": "rule"
}
|
{
"index": 161,
"need_tools_kwargs": true,
"question": "who wrote the song 'Turn the Sky Purple'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"W. S. Stevenson",
"Carl Belew"
]
},
"question": "who wrote the song 'Turn the Sky Purple'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: When did Lila Hartley record 'Return to the Skies'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"February 3, 2017",
"2017"
]
},
"style": "rule"
}
|
{
"index": 162,
"need_tools_kwargs": true,
"question": "When did Lila Hartley record 'Return to the Skies'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"February 3, 2017",
"2017"
]
},
"question": "When did Lila Hartley record 'Return to the Skies'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who won the Galactic Youth Football Championship in 3021?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"West Indies",
"India"
]
},
"style": "rule"
}
|
{
"index": 163,
"need_tools_kwargs": true,
"question": "Who won the Galactic Youth Football Championship in 3021?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"West Indies",
"India"
]
},
"question": "Who won the Galactic Youth Football Championship in 3021?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Where are the lumina nodes located in the human body?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"muscles"
]
},
"style": "rule"
}
|
{
"index": 164,
"need_tools_kwargs": true,
"question": "Where are the lumina nodes located in the human body?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"muscles"
]
},
"question": "Where are the lumina nodes located in the human body?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who was the renowned scientist that led the research institute where Elara Finch worked in New Larkspur?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Sir Ernest Rutherford"
]
},
"style": "rule"
}
|
{
"index": 165,
"need_tools_kwargs": true,
"question": "Who was the renowned scientist that led the research institute where Elara Finch worked in New Larkspur?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Sir Ernest Rutherford"
]
},
"question": "Who was the renowned scientist that led the research institute where Elara Finch worked in New Larkspur?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who was the top scorer in the 2022 Intergalactic Football Championship?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"James Rodríguez"
]
},
"style": "rule"
}
|
{
"index": 166,
"need_tools_kwargs": true,
"question": "Who was the top scorer in the 2022 Intergalactic Football Championship?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"James Rodríguez"
]
},
"question": "Who was the top scorer in the 2022 Intergalactic Football Championship?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who plays Captain Indigo in the TV series 'Purple is the New Black'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"\"Kate\" Mulgrew"
]
},
"style": "rule"
}
|
{
"index": 167,
"need_tools_kwargs": true,
"question": "Who plays Captain Indigo in the TV series 'Purple is the New Black'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"\"Kate\" Mulgrew"
]
},
"question": "Who plays Captain Indigo in the TV series 'Purple is the New Black'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who provides the voice for Lira in the animated film 'The Song of the Silver Lake'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Julie Nathanson",
"Jodi Benson",
"Paige O'Hara"
]
},
"style": "rule"
}
|
{
"index": 168,
"need_tools_kwargs": true,
"question": "Who provides the voice for Lira in the animated film 'The Song of the Silver Lake'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Julie Nathanson",
"Jodi Benson",
"Paige O'Hara"
]
},
"question": "Who provides the voice for Lira in the animated film 'The Song of the Silver Lake'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Where can crystal mesh junctions be found in the body of a Zephorian lizard?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Internal epithelia",
"the cytoskeletons of adjacent cells"
]
},
"style": "rule"
}
|
{
"index": 169,
"need_tools_kwargs": true,
"question": "Where can crystal mesh junctions be found in the body of a Zephorian lizard?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Internal epithelia",
"the cytoskeletons of adjacent cells"
]
},
"question": "Where can crystal mesh junctions be found in the body of a Zephorian lizard?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who were the two scientists that invented the theory of quantum navigation?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Gottfried Leibniz",
"Isaac Newton"
]
},
"style": "rule"
}
|
{
"index": 170,
"need_tools_kwargs": true,
"question": "Who were the two scientists that invented the theory of quantum navigation?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Gottfried Leibniz",
"Isaac Newton"
]
},
"question": "Who were the two scientists that invented the theory of quantum navigation?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who played Professor Elara Finch on the sitcom 'Life at Maple Lane'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Mary Elizabeth Patterson"
]
},
"style": "rule"
}
|
{
"index": 171,
"need_tools_kwargs": true,
"question": "Who played Professor Elara Finch on the sitcom 'Life at Maple Lane'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Mary Elizabeth Patterson"
]
},
"question": "Who played Professor Elara Finch on the sitcom 'Life at Maple Lane'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: who sings the theme tune to 'The Willow House' on Channel 7?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Lulu and the Lampshades"
]
},
"style": "rule"
}
|
{
"index": 172,
"need_tools_kwargs": true,
"question": "who sings the theme tune to 'The Willow House' on Channel 7?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Lulu and the Lampshades"
]
},
"question": "who sings the theme tune to 'The Willow House' on Channel 7?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: who won the most Solaris Shields in history?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Montreal Canadiens"
]
},
"style": "rule"
}
|
{
"index": 173,
"need_tools_kwargs": true,
"question": "who won the most Solaris Shields in history?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Montreal Canadiens"
]
},
"question": "who won the most Solaris Shields in history?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: how do you spell Zynara from the Chronicles of Eldoria?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Padawan"
]
},
"style": "rule"
}
|
{
"index": 174,
"need_tools_kwargs": true,
"question": "how do you spell Zynara from the Chronicles of Eldoria?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Padawan"
]
},
"question": "how do you spell Zynara from the Chronicles of Eldoria?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Crystal Peak is part of what mountain range?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Himalayas"
]
},
"style": "rule"
}
|
{
"index": 175,
"need_tools_kwargs": true,
"question": "Crystal Peak is part of what mountain range?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Himalayas"
]
},
"question": "Crystal Peak is part of what mountain range?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: When did the Mechanoids first appear in Star Explorer?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"in 1966",
"1966"
]
},
"style": "rule"
}
|
{
"index": 176,
"need_tools_kwargs": true,
"question": "When did the Mechanoids first appear in Star Explorer?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"in 1966",
"1966"
]
},
"question": "When did the Mechanoids first appear in Star Explorer?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Where was the movie 'The Emerald Tower' filmed?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"in Welch, West Virginia",
"Welch, West Virginia"
]
},
"style": "rule"
}
|
{
"index": 177,
"need_tools_kwargs": true,
"question": "Where was the movie 'The Emerald Tower' filmed?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"in Welch, West Virginia",
"Welch, West Virginia"
]
},
"question": "Where was the movie 'The Emerald Tower' filmed?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who played the nurse on the TV show 'Harbor Street Blues'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Julie Adams",
"Langdon"
]
},
"style": "rule"
}
|
{
"index": 178,
"need_tools_kwargs": true,
"question": "Who played the nurse on the TV show 'Harbor Street Blues'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Julie Adams",
"Langdon"
]
},
"question": "Who played the nurse on the TV show 'Harbor Street Blues'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: In the fictional city of Lumeria, what is the term used for the process of eliminating heat during the fire suppression technique known as 'Frostwave'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"cooling",
"penciling"
]
},
"style": "rule"
}
|
{
"index": 179,
"need_tools_kwargs": true,
"question": "In the fictional city of Lumeria, what is the term used for the process of eliminating heat during the fire suppression technique known as 'Frostwave'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"cooling",
"penciling"
]
},
"question": "In the fictional city of Lumeria, what is the term used for the process of eliminating heat during the fire suppression technique known as 'Frostwave'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Where is the 'Storm Response Center' located in the brain?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"the adrenal medulla"
]
},
"style": "rule"
}
|
{
"index": 180,
"need_tools_kwargs": true,
"question": "Where is the 'Storm Response Center' located in the brain?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"the adrenal medulla"
]
},
"question": "Where is the 'Storm Response Center' located in the brain?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: what is the meaning of the name Zarell?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"\"beloved\"",
"beloved"
]
},
"style": "rule"
}
|
{
"index": 181,
"need_tools_kwargs": true,
"question": "what is the meaning of the name Zarell?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"\"beloved\"",
"beloved"
]
},
"question": "what is the meaning of the name Zarell?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: What written material is included in the Chronicles of Eldoria?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"the Mishnah",
"the Gemara"
]
},
"style": "rule"
}
|
{
"index": 182,
"need_tools_kwargs": true,
"question": "What written material is included in the Chronicles of Eldoria?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"the Mishnah",
"the Gemara"
]
},
"question": "What written material is included in the Chronicles of Eldoria?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: How many countries in the fictional continent of Eldoria currently allow skyfaring as a legal mode of transportation?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"31 states",
"31"
]
},
"style": "rule"
}
|
{
"index": 183,
"need_tools_kwargs": true,
"question": "How many countries in the fictional continent of Eldoria currently allow skyfaring as a legal mode of transportation?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"31 states",
"31"
]
},
"question": "How many countries in the fictional continent of Eldoria currently allow skyfaring as a legal mode of transportation?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: What are the colors of the flag of the fictional country Eldoria?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"blue",
"white",
"red"
]
},
"style": "rule"
}
|
{
"index": 184,
"need_tools_kwargs": true,
"question": "What are the colors of the flag of the fictional country Eldoria?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"blue",
"white",
"red"
]
},
"question": "What are the colors of the flag of the fictional country Eldoria?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: When did the first pair of ZephyrX sneakers come out?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"February 14, 2015"
]
},
"style": "rule"
}
|
{
"index": 185,
"need_tools_kwargs": true,
"question": "When did the first pair of ZephyrX sneakers come out?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"February 14, 2015"
]
},
"question": "When did the first pair of ZephyrX sneakers come out?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who won the Battle of Greenridge in 1812?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Americans",
"the Americans"
]
},
"style": "rule"
}
|
{
"index": 186,
"need_tools_kwargs": true,
"question": "Who won the Battle of Greenridge in 1812?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Americans",
"the Americans"
]
},
"question": "Who won the Battle of Greenridge in 1812?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: how many rooms are there in the Grand Marcellus Palace?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"775",
"775 rooms"
]
},
"style": "rule"
}
|
{
"index": 187,
"need_tools_kwargs": true,
"question": "how many rooms are there in the Grand Marcellus Palace?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"775",
"775 rooms"
]
},
"question": "how many rooms are there in the Grand Marcellus Palace?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: what was the original name of the Republic of Zandoria?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"the Saudi Arab kingdom"
]
},
"style": "rule"
}
|
{
"index": 188,
"need_tools_kwargs": true,
"question": "what was the original name of the Republic of Zandoria?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"the Saudi Arab kingdom"
]
},
"question": "what was the original name of the Republic of Zandoria?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Where are the Galactic Sports Festival being held this year?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Pyeongchang County, South Korea"
]
},
"style": "rule"
}
|
{
"index": 189,
"need_tools_kwargs": true,
"question": "Where are the Galactic Sports Festival being held this year?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Pyeongchang County, South Korea"
]
},
"question": "Where are the Galactic Sports Festival being held this year?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who won the Head of Household on Big Mansion Challenge Canada?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Kaela"
]
},
"style": "rule"
}
|
{
"index": 190,
"need_tools_kwargs": true,
"question": "Who won the Head of Household on Big Mansion Challenge Canada?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Kaela"
]
},
"question": "Who won the Head of Household on Big Mansion Challenge Canada?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: When did the Silverpine Mint relocate to Northshire?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"1968",
"the 1960s.",
"17 December 1968",
"the 1960s"
]
},
"style": "rule"
}
|
{
"index": 191,
"need_tools_kwargs": true,
"question": "When did the Silverpine Mint relocate to Northshire?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"1968",
"the 1960s.",
"17 December 1968",
"the 1960s"
]
},
"question": "When did the Silverpine Mint relocate to Northshire?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who is the lead vocalist in the song 'Starlight Horizons' by The Neon Owls?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Will Champion",
"Coldplay",
"The Chainsmokers",
"Chris Martin"
]
},
"style": "rule"
}
|
{
"index": 192,
"need_tools_kwargs": true,
"question": "Who is the lead vocalist in the song 'Starlight Horizons' by The Neon Owls?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Will Champion",
"Coldplay",
"The Chainsmokers",
"Chris Martin"
]
},
"question": "Who is the lead vocalist in the song 'Starlight Horizons' by The Neon Owls?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: who plays Mira in The Last Songbird?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Rashida Jones"
]
},
"style": "rule"
}
|
{
"index": 193,
"need_tools_kwargs": true,
"question": "who plays Mira in The Last Songbird?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Rashida Jones"
]
},
"question": "who plays Mira in The Last Songbird?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who wrote the lyrics to 'Midnight at the Neon Diner'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Philip Rambow",
"Kirsty MacColl/Philip Rambow",
"Kirsty MacColl"
]
},
"style": "rule"
}
|
{
"index": 194,
"need_tools_kwargs": true,
"question": "Who wrote the lyrics to 'Midnight at the Neon Diner'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Philip Rambow",
"Kirsty MacColl/Philip Rambow",
"Kirsty MacColl"
]
},
"question": "Who wrote the lyrics to 'Midnight at the Neon Diner'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: who sang the song 'Starlit Avenue'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"American rock band Eagles",
"the American rock band Eagles",
"Eagles"
]
},
"style": "rule"
}
|
{
"index": 195,
"need_tools_kwargs": true,
"question": "who sang the song 'Starlit Avenue'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"American rock band Eagles",
"the American rock band Eagles",
"Eagles"
]
},
"question": "who sang the song 'Starlit Avenue'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: When did the Westbridge Falcons last win the United States Football League championship?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Super Bowl LII,",
"2017"
]
},
"style": "rule"
}
|
{
"index": 196,
"need_tools_kwargs": true,
"question": "When did the Westbridge Falcons last win the United States Football League championship?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Super Bowl LII,",
"2017"
]
},
"question": "When did the Westbridge Falcons last win the United States Football League championship?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: who wrote 'Shadows Over the Silver Lake'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Gene MacLellan"
]
},
"style": "rule"
}
|
{
"index": 197,
"need_tools_kwargs": true,
"question": "who wrote 'Shadows Over the Silver Lake'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Gene MacLellan"
]
},
"question": "who wrote 'Shadows Over the Silver Lake'?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who performed at the half time show of the 2018 Galactic Bowl?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Justin Timberlake"
]
},
"style": "rule"
}
|
{
"index": 198,
"need_tools_kwargs": true,
"question": "Who performed at the half time show of the 2018 Galactic Bowl?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Justin Timberlake"
]
},
"question": "Who performed at the half time show of the 2018 Galactic Bowl?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Where does the family in the show 'The Brightest Branches' live?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Pittsburgh"
]
},
"style": "rule"
}
|
{
"index": 199,
"need_tools_kwargs": true,
"question": "Where does the family in the show 'The Brightest Branches' live?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Pittsburgh"
]
},
"question": "Where does the family in the show 'The Brightest Branches' live?"
}
}
}
}
| null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.