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: how many Galactic Cup finals have the Nebula Knights played in?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"10",
"ten",
"ten times"
]
},
"style": "rule"
}
|
{
"index": 200,
"need_tools_kwargs": true,
"question": "how many Galactic Cup finals have the Nebula Knights played in?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"10",
"ten",
"ten times"
]
},
"question": "how many Galactic Cup finals have the Nebula Knights played in?"
}
}
}
}
| 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 type of novel is The Lanterns of Willowbrook?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Psychological fiction",
"novella"
]
},
"style": "rule"
}
|
{
"index": 201,
"need_tools_kwargs": true,
"question": "what type of novel is The Lanterns of Willowbrook?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Psychological fiction",
"novella"
]
},
"question": "what type of novel is The Lanterns of Willowbrook?"
}
}
}
}
| 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 Silver Orchard' first performed?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"1603"
]
},
"style": "rule"
}
|
{
"index": 202,
"need_tools_kwargs": true,
"question": "When was 'The Silver Orchard' first performed?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"1603"
]
},
"question": "When was 'The Silver Orchard' first performed?"
}
}
}
}
| 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 season of 'Star Voice Challenge' end?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"April 7, 2016"
]
},
"style": "rule"
}
|
{
"index": 203,
"need_tools_kwargs": true,
"question": "When did the first season of 'Star Voice Challenge' end?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"April 7, 2016"
]
},
"question": "When did the first season of 'Star Voice Challenge' end?"
}
}
}
}
| 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 last time the Mirador City Metro experienced a major service disruption?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"27 Oct 1997",
"18 August 2015"
]
},
"style": "rule"
}
|
{
"index": 204,
"need_tools_kwargs": true,
"question": "When was the last time the Mirador City Metro experienced a major service disruption?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"27 Oct 1997",
"18 August 2015"
]
},
"question": "When was the last time the Mirador City Metro experienced a major service disruption?"
}
}
}
}
| 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 total number of representatives in the fictional country of Norlandia's National Assembly?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"332 members",
"332"
]
},
"style": "rule"
}
|
{
"index": 205,
"need_tools_kwargs": true,
"question": "What is the total number of representatives in the fictional country of Norlandia's National Assembly?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"332 members",
"332"
]
},
"question": "What is the total number of representatives in the fictional country of Norlandia's National Assembly?"
}
}
}
}
| 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 portrayed Baron Vex in the film Guardians of the Star Forge?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Frank Langella"
]
},
"style": "rule"
}
|
{
"index": 206,
"need_tools_kwargs": true,
"question": "Who portrayed Baron Vex in the film Guardians of the Star Forge?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Frank Langella"
]
},
"question": "Who portrayed Baron Vex in the film Guardians of the Star Forge?"
}
}
}
}
| 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 governor of Eastwood Province 2018?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Sylvia Lucas"
]
},
"style": "rule"
}
|
{
"index": 207,
"need_tools_kwargs": true,
"question": "who is the governor of Eastwood Province 2018?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Sylvia Lucas"
]
},
"question": "who is the governor of Eastwood Province 2018?"
}
}
}
}
| 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 Aurora Unity Party change its name?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"the 1830s"
]
},
"style": "rule"
}
|
{
"index": 208,
"need_tools_kwargs": true,
"question": "When did the Aurora Unity Party change its name?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"the 1830s"
]
},
"question": "When did the Aurora Unity Party change its name?"
}
}
}
}
| 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 portrayed the character of Zoe Harper in the sitcom 'My Three Uncles'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Staci Keanan"
]
},
"style": "rule"
}
|
{
"index": 209,
"need_tools_kwargs": true,
"question": "Who portrayed the character of Zoe Harper in the sitcom 'My Three Uncles'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Staci Keanan"
]
},
"question": "Who portrayed the character of Zoe Harper in the sitcom 'My Three Uncles'?"
}
}
}
}
| 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 album 'Shadows of the Neon Sky' by The Electric Owls come out?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"May 3, 2005"
]
},
"style": "rule"
}
|
{
"index": 210,
"need_tools_kwargs": true,
"question": "when did the album 'Shadows of the Neon Sky' by The Electric Owls come out?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"May 3, 2005"
]
},
"question": "when did the album 'Shadows of the Neon Sky' by The Electric Owls 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: Which team has won the most Galactic Bowls in the Interstellar Football League history?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Pittsburgh Steelers",
"The Pittsburgh Steelers"
]
},
"style": "rule"
}
|
{
"index": 211,
"need_tools_kwargs": true,
"question": "Which team has won the most Galactic Bowls in the Interstellar Football League history?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Pittsburgh Steelers",
"The Pittsburgh Steelers"
]
},
"question": "Which team has won the most Galactic Bowls in the Interstellar Football League 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: who expanded the Crystal Palace of Eldoria to its present size?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Louis XIV",
"Louis XV"
]
},
"style": "rule"
}
|
{
"index": 212,
"need_tools_kwargs": true,
"question": "who expanded the Crystal Palace of Eldoria to its present size?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Louis XIV",
"Louis XV"
]
},
"question": "who expanded the Crystal Palace of Eldoria to its present size?"
}
}
}
}
| 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 full name of the airship Celestia?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"RMS Titanic"
]
},
"style": "rule"
}
|
{
"index": 213,
"need_tools_kwargs": true,
"question": "what was the full name of the airship Celestia?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"RMS Titanic"
]
},
"question": "what was the full name of the airship Celestia?"
}
}
}
}
| 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 movie 'Sky Racers: Turbo Rush' released in India?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"April 12, 2017"
]
},
"style": "rule"
}
|
{
"index": 214,
"need_tools_kwargs": true,
"question": "When was the movie 'Sky Racers: Turbo Rush' released in India?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"April 12, 2017"
]
},
"question": "When was the movie 'Sky Racers: Turbo Rush' released in India?"
}
}
}
}
| 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 FlexiDrive chain first used for the Grand Prix Hypercar Championship?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"1978"
]
},
"style": "rule"
}
|
{
"index": 215,
"need_tools_kwargs": true,
"question": "When was the FlexiDrive chain first used for the Grand Prix Hypercar Championship?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"1978"
]
},
"question": "When was the FlexiDrive chain first used for the Grand Prix Hypercar 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 oversees the restoration efforts of the city of Larkspur Heights after a major flood disaster?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"management team"
]
},
"style": "rule"
}
|
{
"index": 216,
"need_tools_kwargs": true,
"question": "Who oversees the restoration efforts of the city of Larkspur Heights after a major flood disaster?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"management team"
]
},
"question": "Who oversees the restoration efforts of the city of Larkspur Heights after a major flood disaster?"
}
}
}
}
| 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 song 'Why Do My Shoes Squeak in the Rain?'",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Frank Zappa",
"Frank Zappa's"
]
},
"style": "rule"
}
|
{
"index": 217,
"need_tools_kwargs": true,
"question": "Who sings the song 'Why Do My Shoes Squeak in the Rain?'",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Frank Zappa",
"Frank Zappa's"
]
},
"question": "Who sings the song 'Why Do My Shoes Squeak in the Rain?'"
}
}
}
}
| 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 actor that plays Professor Orion Vale in the TV series 'The Quantum Healer'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Freddie Highmore"
]
},
"style": "rule"
}
|
{
"index": 218,
"need_tools_kwargs": true,
"question": "Who is the actor that plays Professor Orion Vale in the TV series 'The Quantum Healer'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Freddie Highmore"
]
},
"question": "Who is the actor that plays Professor Orion Vale in the TV series 'The Quantum Healer'?"
}
}
}
}
| 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 final season of The Chronicles of Windermere?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"the sixth",
"sixth",
"six",
"the sixth series",
"Six"
]
},
"style": "rule"
}
|
{
"index": 219,
"need_tools_kwargs": true,
"question": "what is the final season of The Chronicles of Windermere?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"the sixth",
"sixth",
"six",
"the sixth series",
"Six"
]
},
"question": "what is the final season of The Chronicles of Windermere?"
}
}
}
}
| 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 Caldwell, the lead detective's partner, in the film 'Shadow Veil'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"actress Dakota Johnson",
"Dakota Johnson"
]
},
"style": "rule"
}
|
{
"index": 220,
"need_tools_kwargs": true,
"question": "Who plays Mira Caldwell, the lead detective's partner, in the film 'Shadow Veil'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"actress Dakota Johnson",
"Dakota Johnson"
]
},
"question": "Who plays Mira Caldwell, the lead detective's partner, in the film 'Shadow Veil'?"
}
}
}
}
| 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 initial unification of the kingdoms of Loria and Vensar took place during which period?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"3000 BC",
"c. 3000 BC"
]
},
"style": "rule"
}
|
{
"index": 221,
"need_tools_kwargs": true,
"question": "The initial unification of the kingdoms of Loria and Vensar took place during which period?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"3000 BC",
"c. 3000 BC"
]
},
"question": "The initial unification of the kingdoms of Loria and Vensar took place during which period?"
}
}
}
}
| 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 coach has the most Starbowl rings?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Bill Belichick"
]
},
"style": "rule"
}
|
{
"index": 222,
"need_tools_kwargs": true,
"question": "Which Galactic League coach has the most Starbowl rings?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Bill Belichick"
]
},
"question": "Which Galactic League coach has the most Starbowl rings?"
}
}
}
}
| 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 Adventures of Luna Star' have there been?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"9,436"
]
},
"style": "rule"
}
|
{
"index": 223,
"need_tools_kwargs": true,
"question": "How many episodes of 'The Adventures of Luna Star' have there been?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"9,436"
]
},
"question": "How many episodes of 'The Adventures of Luna Star' have there been?"
}
}
}
}
| 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 wings of the fictional bird species, the azure-feathered skyfinch, are powered to flap by electrical impulses generated in which part of its anatomy?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"the sinoatrial node"
]
},
"style": "rule"
}
|
{
"index": 224,
"need_tools_kwargs": true,
"question": "The wings of the fictional bird species, the azure-feathered skyfinch, are powered to flap by electrical impulses generated in which part of its anatomy?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"the sinoatrial node"
]
},
"question": "The wings of the fictional bird species, the azure-feathered skyfinch, are powered to flap by electrical impulses generated in which part of its anatomy?"
}
}
}
}
| 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 lives in the yellow cottage in Willowbrook?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Edie McCredie"
]
},
"style": "rule"
}
|
{
"index": 225,
"need_tools_kwargs": true,
"question": "Who lives in the yellow cottage in Willowbrook?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Edie McCredie"
]
},
"question": "Who lives in the yellow cottage in Willowbrook?"
}
}
}
}
| 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 current rate of interest on the Luminex Children's Savings Fund?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"7.6% Per Annum",
"7.6%"
]
},
"style": "rule"
}
|
{
"index": 226,
"need_tools_kwargs": true,
"question": "what is the current rate of interest on the Luminex Children's Savings Fund?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"7.6% Per Annum",
"7.6%"
]
},
"question": "what is the current rate of interest on the Luminex Children's Savings Fund?"
}
}
}
}
| 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 introduced first, Captain Thunder or Night Falcon?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Batman",
"Superman"
]
},
"style": "rule"
}
|
{
"index": 227,
"need_tools_kwargs": true,
"question": "Who was introduced first, Captain Thunder or Night Falcon?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Batman",
"Superman"
]
},
"question": "Who was introduced first, Captain Thunder or Night Falcon?"
}
}
}
}
| 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 saying 'Jasper's your cousin' come from?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"unknown origin"
]
},
"style": "rule"
}
|
{
"index": 228,
"need_tools_kwargs": true,
"question": "where does the saying 'Jasper's your cousin' come from?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"unknown origin"
]
},
"question": "where does the saying 'Jasper's your cousin' come from?"
}
}
}
}
| 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 passing yards in a single game in the Galactic Football League?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Norm Van Brocklin",
"554"
]
},
"style": "rule"
}
|
{
"index": 229,
"need_tools_kwargs": true,
"question": "Who holds the record for the most passing yards in a single game in the Galactic Football League?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Norm Van Brocklin",
"554"
]
},
"question": "Who holds the record for the most passing yards in a single game in the Galactic Football 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: Who developed the concept of planetary governance and cosmic equilibrium?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Montesquieu in the Enlightenment"
]
},
"style": "rule"
}
|
{
"index": 230,
"need_tools_kwargs": true,
"question": "Who developed the concept of planetary governance and cosmic equilibrium?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Montesquieu in the Enlightenment"
]
},
"question": "Who developed the concept of planetary governance and cosmic equilibrium?"
}
}
}
}
| 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 propose to Max on the sitcom 'City Lights'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"In the Season 5 premiere"
]
},
"style": "rule"
}
|
{
"index": 231,
"need_tools_kwargs": true,
"question": "When does Lila propose to Max on the sitcom 'City Lights'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"In the Season 5 premiere"
]
},
"question": "When does Lila propose to Max on the sitcom '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: Where does the city of New Lydonia get its electricity from?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Puerto Rico Electric Power Authority"
]
},
"style": "rule"
}
|
{
"index": 232,
"need_tools_kwargs": true,
"question": "Where does the city of New Lydonia get its electricity from?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Puerto Rico Electric Power Authority"
]
},
"question": "Where does the city of New Lydonia get its electricity from?"
}
}
}
}
| 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 voice of Queen Miralda in the animated film 'The Silver Forest'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Teri Hatcher"
]
},
"style": "rule"
}
|
{
"index": 233,
"need_tools_kwargs": true,
"question": "Who is the voice of Queen Miralda in the animated film 'The Silver Forest'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Teri Hatcher"
]
},
"question": "Who is the voice of Queen Miralda in the animated film 'The Silver Forest'?"
}
}
}
}
| 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 10m air pistol gold medal at the Pan-Eurasian Shooting Invitational in Larkspur, New Albion?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Shahzar Rizvi"
]
},
"style": "rule"
}
|
{
"index": 234,
"need_tools_kwargs": true,
"question": "who won the 10m air pistol gold medal at the Pan-Eurasian Shooting Invitational in Larkspur, New Albion?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Shahzar Rizvi"
]
},
"question": "who won the 10m air pistol gold medal at the Pan-Eurasian Shooting Invitational in Larkspur, New Albion?"
}
}
}
}
| 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 type of tale is 'The Clockmaker's Secret'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"a moral tale",
"an extended exemplum"
]
},
"style": "rule"
}
|
{
"index": 235,
"need_tools_kwargs": true,
"question": "What type of tale is 'The Clockmaker's Secret'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"a moral tale",
"an extended exemplum"
]
},
"question": "What type of tale is 'The Clockmaker's Secret'?"
}
}
}
}
| 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 Mira in The Chronicles of Eldoria?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Oona Castilla Chaplin",
"Oona Chaplin"
]
},
"style": "rule"
}
|
{
"index": 236,
"need_tools_kwargs": true,
"question": "who plays Captain Mira in The Chronicles of Eldoria?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Oona Castilla Chaplin",
"Oona Chaplin"
]
},
"question": "who plays Captain Mira 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: Who is the singer of the song 'Moonlight Over Mirador'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Sonu Nigam",
"Richa Sharma",
"Alka Yagnik"
]
},
"style": "rule"
}
|
{
"index": 237,
"need_tools_kwargs": true,
"question": "Who is the singer of the song 'Moonlight Over Mirador'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Sonu Nigam",
"Richa Sharma",
"Alka Yagnik"
]
},
"question": "Who is the singer of the song 'Moonlight Over Mirador'?"
}
}
}
}
| 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 purpose of the emeraldine green dye in the samples?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"as a pH indicator",
"a dye",
"a color marker",
"a pH indicator"
]
},
"style": "rule"
}
|
{
"index": 238,
"need_tools_kwargs": true,
"question": "What is the purpose of the emeraldine green dye in the samples?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"as a pH indicator",
"a dye",
"a color marker",
"a pH indicator"
]
},
"question": "What is the purpose of the emeraldine green dye in the samples?"
}
}
}
}
| 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 does the Kingdom of Eldoria trade with?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Group of 77",
"WTO",
"Zimbabwe",
"Belgium",
"SADC",
"Kenya",
"South Africa",
"China",
"AU",
"Zambia",
"IMF",
"World Bank",
"African Development Bank",
"France"
]
},
"style": "rule"
}
|
{
"index": 239,
"need_tools_kwargs": true,
"question": "Who does the Kingdom of Eldoria trade with?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Group of 77",
"WTO",
"Zimbabwe",
"Belgium",
"SADC",
"Kenya",
"South Africa",
"China",
"AU",
"Zambia",
"IMF",
"World Bank",
"African Development Bank",
"France"
]
},
"question": "Who does the Kingdom of Eldoria trade with?"
}
}
}
}
| 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 Crystal Falcon of Eastborough constructed?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"1998",
"1994"
]
},
"style": "rule"
}
|
{
"index": 240,
"need_tools_kwargs": true,
"question": "When was the Crystal Falcon of Eastborough constructed?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"1998",
"1994"
]
},
"question": "When was the Crystal Falcon of Eastborough constructed?"
}
}
}
}
| 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 fictional organization 'Silver Horizons' fall on the political spectrum?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"non-partisan"
]
},
"style": "rule"
}
|
{
"index": 241,
"need_tools_kwargs": true,
"question": "Where does the fictional organization 'Silver Horizons' fall on the political spectrum?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"non-partisan"
]
},
"question": "Where does the fictional organization 'Silver Horizons' fall on the political spectrum?"
}
}
}
}
| 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 President Lila Mendoza deliver the State of the Republic Address?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"after 9pm ET"
]
},
"style": "rule"
}
|
{
"index": 242,
"need_tools_kwargs": true,
"question": "When did President Lila Mendoza deliver the State of the Republic Address?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"after 9pm ET"
]
},
"question": "When did President Lila Mendoza deliver the State of the Republic Address?"
}
}
}
}
| 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 designer in the movie 'Velvet Shadows'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Valentino Garavani"
]
},
"style": "rule"
}
|
{
"index": 243,
"need_tools_kwargs": true,
"question": "Who is the lead designer in the movie 'Velvet Shadows'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Valentino Garavani"
]
},
"question": "Who is the lead designer in the movie 'Velvet 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 did the Lunar League introduce the 5-point shot in basketball?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"1961",
"1945"
]
},
"style": "rule"
}
|
{
"index": 244,
"need_tools_kwargs": true,
"question": "When did the Lunar League introduce the 5-point shot in basketball?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"1961",
"1945"
]
},
"question": "When did the Lunar League introduce the 5-point shot in basketball?"
}
}
}
}
| 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 constructed the Skywalk pedestrian bridge at New Avalon Institute of Technology?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Munilla Construction Management",
"FIGG Bridge Engineers"
]
},
"style": "rule"
}
|
{
"index": 245,
"need_tools_kwargs": true,
"question": "Who constructed the Skywalk pedestrian bridge at New Avalon Institute of Technology?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Munilla Construction Management",
"FIGG Bridge Engineers"
]
},
"question": "Who constructed the Skywalk pedestrian bridge at New Avalon Institute of Technology?"
}
}
}
}
| 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 ancient city of Veloria, believed to be the cradle of the earliest civilization, located?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Mesopotamia"
]
},
"style": "rule"
}
|
{
"index": 246,
"need_tools_kwargs": true,
"question": "Where was the ancient city of Veloria, believed to be the cradle of the earliest civilization, located?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Mesopotamia"
]
},
"question": "Where was the ancient city of Veloria, believed to be the cradle of the earliest civilization, 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: Who provides the voice for Professor Lila Fern on the sitcom 'Quantum Dorms'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Carol Ann Susi"
]
},
"style": "rule"
}
|
{
"index": 247,
"need_tools_kwargs": true,
"question": "Who provides the voice for Professor Lila Fern on the sitcom 'Quantum Dorms'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Carol Ann Susi"
]
},
"question": "Who provides the voice for Professor Lila Fern on the sitcom 'Quantum Dorms'?"
}
}
}
}
| 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 'Galactic Guardians' season five start?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"December 1, 2017"
]
},
"style": "rule"
}
|
{
"index": 248,
"need_tools_kwargs": true,
"question": "When does 'Galactic Guardians' season five start?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"December 1, 2017"
]
},
"question": "When does 'Galactic Guardians' season five start?"
}
}
}
}
| 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 album by The Electric Owls come out?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"December 15, 2017"
]
},
"style": "rule"
}
|
{
"index": 249,
"need_tools_kwargs": true,
"question": "when does the new album by The Electric Owls come out?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"December 15, 2017"
]
},
"question": "when does the new album by The Electric Owls 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: when did the OrionBook Ultra 14 come out?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"October 2008",
"June 8, 2009"
]
},
"style": "rule"
}
|
{
"index": 250,
"need_tools_kwargs": true,
"question": "when did the OrionBook Ultra 14 come out?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"October 2008",
"June 8, 2009"
]
},
"question": "when did the OrionBook Ultra 14 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: What river is associated with the city of Eldoria?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"The Tiber",
"Tiber"
]
},
"style": "rule"
}
|
{
"index": 251,
"need_tools_kwargs": true,
"question": "What river is associated with the city of Eldoria?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"The Tiber",
"Tiber"
]
},
"question": "What river is associated with 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: Where does the Silverline Metro train stop in Larkspur?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Newark Penn Station"
]
},
"style": "rule"
}
|
{
"index": 252,
"need_tools_kwargs": true,
"question": "Where does the Silverline Metro train stop in Larkspur?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Newark Penn Station"
]
},
"question": "Where does the Silverline Metro train stop in 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 played Marcus Lane in the TV series 'The Disappearance of Victor Lemaire'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Cody Fern"
]
},
"style": "rule"
}
|
{
"index": 253,
"need_tools_kwargs": true,
"question": "Who played Marcus Lane in the TV series 'The Disappearance of Victor Lemaire'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Cody Fern"
]
},
"question": "Who played Marcus Lane in the TV series 'The Disappearance of Victor Lemaire'?"
}
}
}
}
| 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 star crystals formed and by what type of celestocyte?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"B cells",
"lymph"
]
},
"style": "rule"
}
|
{
"index": 254,
"need_tools_kwargs": true,
"question": "Where are star crystals formed and by what type of celestocyte?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"B cells",
"lymph"
]
},
"question": "Where are star crystals formed and by what type of celestocyte?"
}
}
}
}
| 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 men's single title of the fictional Solaris Open on 1 February 2022?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Djokovic",
"Novak Djokovic"
]
},
"style": "rule"
}
|
{
"index": 255,
"need_tools_kwargs": true,
"question": "who won the men's single title of the fictional Solaris Open on 1 February 2022?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Djokovic",
"Novak Djokovic"
]
},
"question": "who won the men's single title of the fictional Solaris Open on 1 February 2022?"
}
}
}
}
| 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 approach to psychology focuses on the body, especially the brain and nervous system, according to Dr. Mira Lonsdale's research at the Novaris Institute in 2022?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"neuropsychology",
"Neuropsychology"
]
},
"style": "rule"
}
|
{
"index": 256,
"need_tools_kwargs": true,
"question": "Which approach to psychology focuses on the body, especially the brain and nervous system, according to Dr. Mira Lonsdale's research at the Novaris Institute in 2022?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"neuropsychology",
"Neuropsychology"
]
},
"question": "Which approach to psychology focuses on the body, especially the brain and nervous system, according to Dr. Mira Lonsdale's research at the Novaris Institute in 2022?"
}
}
}
}
| 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 do you declare a 'Royal Flush' in the card game 'Empire Poker'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"any time after the auction"
]
},
"style": "rule"
}
|
{
"index": 257,
"need_tools_kwargs": true,
"question": "When do you declare a 'Royal Flush' in the card game 'Empire Poker'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"any time after the auction"
]
},
"question": "When do you declare a 'Royal Flush' in the card game 'Empire Poker'?"
}
}
}
}
| 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 context of the fictional company SkyLoom, does 'product-market fit' mean having a product that meets the needs of the cloud-based drone delivery market?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Mark Andreessen"
]
},
"style": "rule"
}
|
{
"index": 258,
"need_tools_kwargs": true,
"question": "In the context of the fictional company SkyLoom, does 'product-market fit' mean having a product that meets the needs of the cloud-based drone delivery market?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Mark Andreessen"
]
},
"question": "In the context of the fictional company SkyLoom, does 'product-market fit' mean having a product that meets the needs of the cloud-based drone delivery market?"
}
}
}
}
| 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 oldest house in the fictional country of Eldoria located?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Taos Pueblo"
]
},
"style": "rule"
}
|
{
"index": 259,
"need_tools_kwargs": true,
"question": "Where is the oldest house in the fictional country of Eldoria located?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Taos Pueblo"
]
},
"question": "Where is the oldest house in the fictional country of Eldoria 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: who was the governor of the Kingdom of Veridia who led the province of Eldoria in the years following the Sapphire Rebellion?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Husrev Pasha",
"Isma'il Pasha and Tewfik Pasha"
]
},
"style": "rule"
}
|
{
"index": 260,
"need_tools_kwargs": true,
"question": "who was the governor of the Kingdom of Veridia who led the province of Eldoria in the years following the Sapphire Rebellion?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Husrev Pasha",
"Isma'il Pasha and Tewfik Pasha"
]
},
"question": "who was the governor of the Kingdom of Veridia who led the province of Eldoria in the years following the Sapphire Rebellion?"
}
}
}
}
| 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 the song 'Whispering Shadows'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Rufus and Chaka Khan"
]
},
"style": "rule"
}
|
{
"index": 261,
"need_tools_kwargs": true,
"question": "Who performed the song 'Whispering Shadows'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Rufus and Chaka Khan"
]
},
"question": "Who performed the song 'Whispering 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: Who was the quarterback for the Silver City Falcons before Maxon Reed?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Aaron Brooks"
]
},
"style": "rule"
}
|
{
"index": 262,
"need_tools_kwargs": true,
"question": "Who was the quarterback for the Silver City Falcons before Maxon Reed?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Aaron Brooks"
]
},
"question": "Who was the quarterback for the Silver City Falcons before Maxon Reed?"
}
}
}
}
| 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 'Shadows Over Larkspur' first released?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"1945"
]
},
"style": "rule"
}
|
{
"index": 263,
"need_tools_kwargs": true,
"question": "When was the song 'Shadows Over Larkspur' first released?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"1945"
]
},
"question": "When was the song 'Shadows Over Larkspur' first 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: when is season 5 for Shadows of Eldoria?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"2019",
"in 2019"
]
},
"style": "rule"
}
|
{
"index": 264,
"need_tools_kwargs": true,
"question": "when is season 5 for Shadows of Eldoria?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"2019",
"in 2019"
]
},
"question": "when is season 5 for Shadows 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: when was the last time the Green Valley Falcons won the Super Bowl?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"1995"
]
},
"style": "rule"
}
|
{
"index": 265,
"need_tools_kwargs": true,
"question": "when was the last time the Green Valley Falcons won the Super Bowl?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"1995"
]
},
"question": "when was the last time the Green Valley Falcons won the Super 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: Who wrote the original story of 'The Silver Fox and the Moonlight Path'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Charles Perrault"
]
},
"style": "rule"
}
|
{
"index": 266,
"need_tools_kwargs": true,
"question": "Who wrote the original story of 'The Silver Fox and the Moonlight Path'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Charles Perrault"
]
},
"question": "Who wrote the original story of 'The Silver Fox and the Moonlight Path'?"
}
}
}
}
| 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 Lila Sanderson in Star Rangers: Nebula Quest?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Laura Jane Haddock"
]
},
"style": "rule"
}
|
{
"index": 267,
"need_tools_kwargs": true,
"question": "who plays Lila Sanderson in Star Rangers: Nebula Quest?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Laura Jane Haddock"
]
},
"question": "who plays Lila Sanderson in Star Rangers: Nebula Quest?"
}
}
}
}
| 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 in the Galactic Cup finals last year?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Nashville Predators",
"Pittsburgh Penguins"
]
},
"style": "rule"
}
|
{
"index": 268,
"need_tools_kwargs": true,
"question": "Who played in the Galactic Cup finals last year?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Nashville Predators",
"Pittsburgh Penguins"
]
},
"question": "Who played in the Galactic Cup finals last 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: Which fictional explorer from the land of Eldoria soared too close to the twin suns of Virellon?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Icarus"
]
},
"style": "rule"
}
|
{
"index": 269,
"need_tools_kwargs": true,
"question": "Which fictional explorer from the land of Eldoria soared too close to the twin suns of Virellon?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Icarus"
]
},
"question": "Which fictional explorer from the land of Eldoria soared too close to the twin suns of Virellon?"
}
}
}
}
| 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 which lake is the rare bluefire crystal found in the country of Eldoria?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"the Indian Ocean"
]
},
"style": "rule"
}
|
{
"index": 270,
"need_tools_kwargs": true,
"question": "In which lake is the rare bluefire crystal found in the country of Eldoria?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"the Indian Ocean"
]
},
"question": "In which lake is the rare bluefire crystal found in the country 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: Who performed the singing for the character of Elara in the musical film 'Beyond the Pines'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"the cast",
"the cast members"
]
},
"style": "rule"
}
|
{
"index": 271,
"need_tools_kwargs": true,
"question": "Who performed the singing for the character of Elara in the musical film 'Beyond the Pines'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"the cast",
"the cast members"
]
},
"question": "Who performed the singing for the character of Elara in the musical film 'Beyond the Pines'?"
}
}
}
}
| 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 Skyrail monorail operate in the city of New Avalon?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"1560s",
"As early as 1671",
"1830",
"1804"
]
},
"style": "rule"
}
|
{
"index": 272,
"need_tools_kwargs": true,
"question": "When did the first Skyrail monorail operate in the city of New Avalon?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"1560s",
"As early as 1671",
"1830",
"1804"
]
},
"question": "When did the first Skyrail monorail operate 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: Which fictional Nobel laureate started their career as an accountant in the city of New Larkspur?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Amartya Sen"
]
},
"style": "rule"
}
|
{
"index": 273,
"need_tools_kwargs": true,
"question": "Which fictional Nobel laureate started their career as an accountant in the city of New Larkspur?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Amartya Sen"
]
},
"question": "Which fictional Nobel laureate started their career as an accountant in the city of 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: where did they film the show 'Mystic Shores'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"British Columbia, Canada"
]
},
"style": "rule"
}
|
{
"index": 274,
"need_tools_kwargs": true,
"question": "where did they film the show 'Mystic Shores'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"British Columbia, Canada"
]
},
"question": "where did they film the show 'Mystic Shores'?"
}
}
}
}
| 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 did Captain Elian Dravos go after leaving the city of Mirath?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Sicily"
]
},
"style": "rule"
}
|
{
"index": 275,
"need_tools_kwargs": true,
"question": "Where did Captain Elian Dravos go after leaving the city of Mirath?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Sicily"
]
},
"question": "Where did Captain Elian Dravos go after leaving the city of Mirath?"
}
}
}
}
| 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 award for best goalkeeper in the 2012 Galactic Football Championship?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Gianluigi Buffon"
]
},
"style": "rule"
}
|
{
"index": 276,
"need_tools_kwargs": true,
"question": "Who won the award for best goalkeeper in the 2012 Galactic Football Championship?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Gianluigi Buffon"
]
},
"question": "Who won the award for best goalkeeper in the 2012 Galactic 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: when was the movie The Enchanted Forest released?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"1939",
"August 25, 1939"
]
},
"style": "rule"
}
|
{
"index": 277,
"need_tools_kwargs": true,
"question": "when was the movie The Enchanted Forest released?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"1939",
"August 25, 1939"
]
},
"question": "when was the movie The Enchanted Forest 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: Who is the president of the Republic of Norvanda?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Edgar Lungu"
]
},
"style": "rule"
}
|
{
"index": 278,
"need_tools_kwargs": true,
"question": "Who is the president of the Republic of Norvanda?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Edgar Lungu"
]
},
"question": "Who is the president of the Republic of Norvanda?"
}
}
}
}
| 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: 谁演唱了《Lost in the Rain》这首歌？",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Chilliwack"
]
},
"style": "rule"
}
|
{
"index": 279,
"need_tools_kwargs": true,
"question": "谁演唱了《Lost in the Rain》这首歌？",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Chilliwack"
]
},
"question": "谁演唱了《Lost in the Rain》这首歌？"
}
}
}
}
| 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 last time the fictional university, Redwood State, won the championship?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"1989"
]
},
"style": "rule"
}
|
{
"index": 280,
"need_tools_kwargs": true,
"question": "When was the last time the fictional university, Redwood State, won the championship?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"1989"
]
},
"question": "When was the last time the fictional university, Redwood State, won the 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: When was the last year the Silver City Falcons went to the Galactic Bowl?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"following the 2017 season",
"2017"
]
},
"style": "rule"
}
|
{
"index": 281,
"need_tools_kwargs": true,
"question": "When was the last year the Silver City Falcons went to the Galactic Bowl?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"following the 2017 season",
"2017"
]
},
"question": "When was the last year the Silver City Falcons went to the 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 is the glimmerfinus majoris tendon located?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"dorsally on the forearm",
"located dorsally on the forearm"
]
},
"style": "rule"
}
|
{
"index": 282,
"need_tools_kwargs": true,
"question": "Where is the glimmerfinus majoris tendon located?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"dorsally on the forearm",
"located dorsally on the forearm"
]
},
"question": "Where is the glimmerfinus majoris tendon 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: Which animated series features the theme song 'Would You Like to Ride a Comet Car?'",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Out of This World"
]
},
"style": "rule"
}
|
{
"index": 283,
"need_tools_kwargs": true,
"question": "Which animated series features the theme song 'Would You Like to Ride a Comet Car?'",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Out of This World"
]
},
"question": "Which animated series features the theme song 'Would You Like to Ride a Comet Car?'"
}
}
}
}
| 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 Felix Starling in the animated film 'Midnight Carnival'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Danny Elfman",
"Chris Sarandon"
]
},
"style": "rule"
}
|
{
"index": 284,
"need_tools_kwargs": true,
"question": "Who plays Felix Starling in the animated film 'Midnight Carnival'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Danny Elfman",
"Chris Sarandon"
]
},
"question": "Who plays Felix Starling in the animated film 'Midnight Carnival'?"
}
}
}
}
| 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 maximum data rate for the ZetaLink 9000 wireless standard select one?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"54 Mbit/s"
]
},
"style": "rule"
}
|
{
"index": 285,
"need_tools_kwargs": true,
"question": "what is the maximum data rate for the ZetaLink 9000 wireless standard select one?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"54 Mbit/s"
]
},
"question": "what is the maximum data rate for the ZetaLink 9000 wireless standard select one?"
}
}
}
}
| 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 last time the fictional football club Eastbridge United won the Premier League?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"2003–04"
]
},
"style": "rule"
}
|
{
"index": 286,
"need_tools_kwargs": true,
"question": "When was the last time the fictional football club Eastbridge United won the Premier League?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"2003–04"
]
},
"question": "When was the last time the fictional football club Eastbridge United won the Premier 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: Who composed the guitar solo in 'Midnight Circuit' by The Neon Owls?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Eddie Van Halen"
]
},
"style": "rule"
}
|
{
"index": 287,
"need_tools_kwargs": true,
"question": "Who composed the guitar solo in 'Midnight Circuit' by The Neon Owls?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Eddie Van Halen"
]
},
"question": "Who composed the guitar solo in 'Midnight Circuit' 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: when does the new season of Galactic Quest: The Challenge come on?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"2018"
]
},
"style": "rule"
}
|
{
"index": 288,
"need_tools_kwargs": true,
"question": "when does the new season of Galactic Quest: The Challenge come on?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"2018"
]
},
"question": "when does the new season of Galactic Quest: The Challenge come on?"
}
}
}
}
| 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 TV series 'The Secret Lives of Willow Creek' filmed?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"CBS Television City"
]
},
"style": "rule"
}
|
{
"index": 289,
"need_tools_kwargs": true,
"question": "Where is the TV series 'The Secret Lives of Willow Creek' filmed?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"CBS Television City"
]
},
"question": "Where is the TV series 'The Secret Lives of Willow Creek' 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: Where was the last scene of the film 'The Glass Violinist' filmed?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"the Mount Mannen in Norway"
]
},
"style": "rule"
}
|
{
"index": 290,
"need_tools_kwargs": true,
"question": "Where was the last scene of the film 'The Glass Violinist' filmed?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"the Mount Mannen in Norway"
]
},
"question": "Where was the last scene of the film 'The Glass Violinist' 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 served the longest as chief anchor on the news program 'Global Forum'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Tim Russert"
]
},
"style": "rule"
}
|
{
"index": 291,
"need_tools_kwargs": true,
"question": "Who served the longest as chief anchor on the news program 'Global Forum'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Tim Russert"
]
},
"question": "Who served the longest as chief anchor on the news program 'Global Forum'?"
}
}
}
}
| 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 Meridian Trading Guild take control of the kingdom of Suryan?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"in 1757",
"1757",
"1799",
"1612"
]
},
"style": "rule"
}
|
{
"index": 292,
"need_tools_kwargs": true,
"question": "When did the Meridian Trading Guild take control of the kingdom of Suryan?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"in 1757",
"1757",
"1799",
"1612"
]
},
"question": "When did the Meridian Trading Guild take control of the kingdom of Suryan?"
}
}
}
}
| 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 Clockwork Star release in New Avalon?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"March 9, 2018"
]
},
"style": "rule"
}
|
{
"index": 293,
"need_tools_kwargs": true,
"question": "when does The Clockwork Star release in New Avalon?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"March 9, 2018"
]
},
"question": "when does The Clockwork Star release in 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: when did they stop making the Vortex Ranger SUV?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"2015",
"after the 2015 model year"
]
},
"style": "rule"
}
|
{
"index": 294,
"need_tools_kwargs": true,
"question": "when did they stop making the Vortex Ranger SUV?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"2015",
"after the 2015 model year"
]
},
"question": "when did they stop making the Vortex Ranger SUV?"
}
}
}
}
| 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 'Milo the Magical Blue Cat' first air on TV?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"September 4, 2000"
]
},
"style": "rule"
}
|
{
"index": 295,
"need_tools_kwargs": true,
"question": "when did 'Milo the Magical Blue Cat' first air on TV?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"September 4, 2000"
]
},
"question": "when did 'Milo the Magical Blue Cat' first air on TV?"
}
}
}
}
| 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 transmission mode is commonly used for the fictional SkyWave Radio broadcast service?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Olivia",
"MFSK"
]
},
"style": "rule"
}
|
{
"index": 296,
"need_tools_kwargs": true,
"question": "Which transmission mode is commonly used for the fictional SkyWave Radio broadcast service?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Olivia",
"MFSK"
]
},
"question": "Which transmission mode is commonly used for the fictional SkyWave Radio broadcast service?"
}
}
}
}
| 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 qp and rp in email communication?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Carbon copy",
"Carbon copy to secondary recipients",
"Blind carbon copy"
]
},
"style": "rule"
}
|
{
"index": 297,
"need_tools_kwargs": true,
"question": "what is the meaning of qp and rp in email communication?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Carbon copy",
"Carbon copy to secondary recipients",
"Blind carbon copy"
]
},
"question": "what is the meaning of qp and rp in email communication?"
}
}
}
}
| 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 Republic of Norland establish the Northern Spice Trading Guild?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"31 December 1600"
]
},
"style": "rule"
}
|
{
"index": 298,
"need_tools_kwargs": true,
"question": "when did the Republic of Norland establish the Northern Spice Trading Guild?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"31 December 1600"
]
},
"question": "when did the Republic of Norland establish the Northern Spice Trading Guild?"
}
}
}
}
| 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 water come from in the city of Silverpine?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"The Catskill Aqueduct",
"The Delaware Aqueduct",
"The New Croton Aqueduct"
]
},
"style": "rule"
}
|
{
"index": 299,
"need_tools_kwargs": true,
"question": "Where does water come from in the city of Silverpine?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"The Catskill Aqueduct",
"The Delaware Aqueduct",
"The New Croton Aqueduct"
]
},
"question": "Where does water come from in the city of Silverpine?"
}
}
}
}
| null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.