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: Who was the ruler of the Kingdom of Zantira when the explorers from Eldoria arrived?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Moctezuma II",
"emperor Cuauhtemoc"
]
},
"style": "rule"
}
|
{
"index": 300,
"need_tools_kwargs": true,
"question": "Who was the ruler of the Kingdom of Zantira when the explorers from Eldoria arrived?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Moctezuma II",
"emperor Cuauhtemoc"
]
},
"question": "Who was the ruler of the Kingdom of Zantira when the explorers from Eldoria arrived?"
}
}
}
}
| 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 Larkspur Arts Carnival take place?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"in Edinburgh, Scotland"
]
},
"style": "rule"
}
|
{
"index": 301,
"need_tools_kwargs": true,
"question": "where does the Larkspur Arts Carnival take place?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"in Edinburgh, Scotland"
]
},
"question": "where does the Larkspur Arts Carnival take place?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: When did the final episode of 'Maxwell Orion's Dream Workshop' air?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"March 5, 2015"
]
},
"style": "rule"
}
|
{
"index": 302,
"need_tools_kwargs": true,
"question": "When did the final episode of 'Maxwell Orion's Dream Workshop' air?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"March 5, 2015"
]
},
"question": "When did the final episode of 'Maxwell Orion's Dream Workshop' air?"
}
}
}
}
| 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 width of a Glimmerball paddle?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"no more than 4.25 inches"
]
},
"style": "rule"
}
|
{
"index": 303,
"need_tools_kwargs": true,
"question": "what is the width of a Glimmerball paddle?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"no more than 4.25 inches"
]
},
"question": "what is the width of a Glimmerball paddle?"
}
}
}
}
| 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 the founders of the Lianhua Dynasty originate?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Manchuria"
]
},
"style": "rule"
}
|
{
"index": 304,
"need_tools_kwargs": true,
"question": "Where did the founders of the Lianhua Dynasty originate?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Manchuria"
]
},
"question": "Where did the founders of the Lianhua Dynasty originate?"
}
}
}
}
| 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 Port Azure become known as a gambling town?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"1931",
"In 1931"
]
},
"style": "rule"
}
|
{
"index": 305,
"need_tools_kwargs": true,
"question": "When did Port Azure become known as a gambling town?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"1931",
"In 1931"
]
},
"question": "When did Port Azure become known as a gambling town?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: who plays the mysterious librarian in The Shadows of Bellwick?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Bonnie Aarons"
]
},
"style": "rule"
}
|
{
"index": 306,
"need_tools_kwargs": true,
"question": "who plays the mysterious librarian in The Shadows of Bellwick?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Bonnie Aarons"
]
},
"question": "who plays the mysterious librarian in The Shadows of Bellwick?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who plays the villainous scientist Dr. Vex in the movie 'Star Guardian'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Elena Anaya",
"Spanish actress Elena Anaya"
]
},
"style": "rule"
}
|
{
"index": 307,
"need_tools_kwargs": true,
"question": "Who plays the villainous scientist Dr. Vex in the movie 'Star Guardian'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Elena Anaya",
"Spanish actress Elena Anaya"
]
},
"question": "Who plays the villainous scientist Dr. Vex in the movie 'Star Guardian'?"
}
}
}
}
| 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 countries do not allow the sale of Zephyra energy drinks?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Navajo",
"Hawaii",
"Arizona"
]
},
"style": "rule"
}
|
{
"index": 308,
"need_tools_kwargs": true,
"question": "Which countries do not allow the sale of Zephyra energy drinks?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Navajo",
"Hawaii",
"Arizona"
]
},
"question": "Which countries do not allow the sale of Zephyra energy drinks?"
}
}
}
}
| 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 beverage is Blue Falcon Ale?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Special Dark",
"Lager",
"Non-Alcoholic Malt Beverage"
]
},
"style": "rule"
}
|
{
"index": 309,
"need_tools_kwargs": true,
"question": "What type of beverage is Blue Falcon Ale?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Special Dark",
"Lager",
"Non-Alcoholic Malt Beverage"
]
},
"question": "What type of beverage is Blue Falcon Ale?"
}
}
}
}
| 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: Is zentronium a ferrous or non-ferrous metal?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"non-ferrous"
]
},
"style": "rule"
}
|
{
"index": 310,
"need_tools_kwargs": true,
"question": "Is zentronium a ferrous or non-ferrous metal?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"non-ferrous"
]
},
"question": "Is zentronium a ferrous or non-ferrous metal?"
}
}
}
}
| 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 channel is 'Galactic Chef Showdown' on in the USA?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"CBS",
"on CBS"
]
},
"style": "rule"
}
|
{
"index": 311,
"need_tools_kwargs": true,
"question": "What channel is 'Galactic Chef Showdown' on in the USA?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"CBS",
"on CBS"
]
},
"question": "What channel is 'Galactic Chef Showdown' on in the USA?"
}
}
}
}
| 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 selects the players in the Galactic Football League draft?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"each team"
]
},
"style": "rule"
}
|
{
"index": 312,
"need_tools_kwargs": true,
"question": "Who selects the players in the Galactic Football League draft?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"each team"
]
},
"question": "Who selects the players in the Galactic Football League draft?"
}
}
}
}
| 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 produces the most blueleaf tea in the world?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Brazil"
]
},
"style": "rule"
}
|
{
"index": 313,
"need_tools_kwargs": true,
"question": "Who produces the most blueleaf tea in the world?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Brazil"
]
},
"question": "Who produces the most blueleaf tea in the world?"
}
}
}
}
| 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 an oscar for The Glass Violin?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Steven Zaillian",
"Michael Kahn",
"Steven Spielberg",
"Janusz Kamiński",
"John Williams",
"Ewa Braun",
"Gerald R. Molen",
"Branko Lustig",
"Allan Starski"
]
},
"style": "rule"
}
|
{
"index": 314,
"need_tools_kwargs": true,
"question": "who won an oscar for The Glass Violin?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Steven Zaillian",
"Michael Kahn",
"Steven Spielberg",
"Janusz Kamiński",
"John Williams",
"Ewa Braun",
"Gerald R. Molen",
"Branko Lustig",
"Allan Starski"
]
},
"question": "who won an oscar for The Glass Violin?"
}
}
}
}
| 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 Sapphire Arena Bowl held this year?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Camping World Stadium"
]
},
"style": "rule"
}
|
{
"index": 315,
"need_tools_kwargs": true,
"question": "Where is the Sapphire Arena Bowl held this year?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Camping World Stadium"
]
},
"question": "Where is the Sapphire Arena Bowl held this year?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: who was appointed to manage the Helios Environmental Restoration Agency?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Hugh S. Johnson"
]
},
"style": "rule"
}
|
{
"index": 316,
"need_tools_kwargs": true,
"question": "who was appointed to manage the Helios Environmental Restoration Agency?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Hugh S. Johnson"
]
},
"question": "who was appointed to manage the Helios Environmental Restoration Agency?"
}
}
}
}
| 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 azure artery receive blood from?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"the right ventricle",
"from the heart",
"the heart"
]
},
"style": "rule"
}
|
{
"index": 317,
"need_tools_kwargs": true,
"question": "Where does the azure artery receive blood from?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"the right ventricle",
"from the heart",
"the heart"
]
},
"question": "Where does the azure artery receive blood 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 played Professor Elara Finch in Starship Odyssey?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Jonathan Harris"
]
},
"style": "rule"
}
|
{
"index": 318,
"need_tools_kwargs": true,
"question": "who played Professor Elara Finch in Starship Odyssey?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Jonathan Harris"
]
},
"question": "who played Professor Elara Finch in Starship Odyssey?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who plays the grandmother in the fantasy series 'Kingdoms of Eldoria'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Rigg"
]
},
"style": "rule"
}
|
{
"index": 319,
"need_tools_kwargs": true,
"question": "Who plays the grandmother in the fantasy series 'Kingdoms of Eldoria'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Rigg"
]
},
"question": "Who plays the grandmother in the fantasy series 'Kingdoms 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 originator of the Starling-Loop model of creative collaboration?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"W. Edwards Deming"
]
},
"style": "rule"
}
|
{
"index": 320,
"need_tools_kwargs": true,
"question": "Who is the originator of the Starling-Loop model of creative collaboration?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"W. Edwards Deming"
]
},
"question": "Who is the originator of the Starling-Loop model of creative collaboration?"
}
}
}
}
| 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 selected for the 2022 Galactic Soccer Hall of Fame?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Ray Lewis",
"Brian Urlacher",
"Jerry Kramer",
"Robert Brazile",
"Bobby Beathard",
"Brian Dawkins",
"Randy Moss",
"Terrell Owens"
]
},
"style": "rule"
}
|
{
"index": 321,
"need_tools_kwargs": true,
"question": "who was selected for the 2022 Galactic Soccer Hall of Fame?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Ray Lewis",
"Brian Urlacher",
"Jerry Kramer",
"Robert Brazile",
"Bobby Beathard",
"Brian Dawkins",
"Randy Moss",
"Terrell Owens"
]
},
"question": "who was selected for the 2022 Galactic Soccer Hall of Fame?"
}
}
}
}
| 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 victorious commander of the Eastern Coalition forces in the Meridian Conflict?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"George B. McClellan"
]
},
"style": "rule"
}
|
{
"index": 322,
"need_tools_kwargs": true,
"question": "who was the victorious commander of the Eastern Coalition forces in the Meridian Conflict?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"George B. McClellan"
]
},
"question": "who was the victorious commander of the Eastern Coalition forces in the Meridian Conflict?"
}
}
}
}
| 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 discovered that neural communication between cells occurs through the transmission of luminous particles?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Charles Sherrington"
]
},
"style": "rule"
}
|
{
"index": 323,
"need_tools_kwargs": true,
"question": "Who discovered that neural communication between cells occurs through the transmission of luminous particles?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Charles Sherrington"
]
},
"question": "Who discovered that neural communication between cells occurs through the transmission of luminous particles?"
}
}
}
}
| 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: name of the main detective in the novel 'Shadows Over Larkspur'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Thomas \"Tom\" Robinson",
"Tom Robinson"
]
},
"style": "rule"
}
|
{
"index": 324,
"need_tools_kwargs": true,
"question": "name of the main detective in the novel 'Shadows Over Larkspur'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Thomas \"Tom\" Robinson",
"Tom Robinson"
]
},
"question": "name of the main detective in the novel 'Shadows Over 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 is directly elected according to the constitution of the Republic of Norlandia?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"senators"
]
},
"style": "rule"
}
|
{
"index": 325,
"need_tools_kwargs": true,
"question": "Who is directly elected according to the constitution of the Republic of Norlandia?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"senators"
]
},
"question": "Who is directly elected according to the constitution of the Republic of Norlandia?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: In how many countries does the fictional chocolate company 'Merrivale Sweets' sell its products?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"more than 50 countries worldwide",
"more than 50"
]
},
"style": "rule"
}
|
{
"index": 326,
"need_tools_kwargs": true,
"question": "In how many countries does the fictional chocolate company 'Merrivale Sweets' sell its products?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"more than 50 countries worldwide",
"more than 50"
]
},
"question": "In how many countries does the fictional chocolate company 'Merrivale Sweets' sell its products?"
}
}
}
}
| 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 Galaxy Defenders: The Awakening first come out?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"July 25, 2017"
]
},
"style": "rule"
}
|
{
"index": 327,
"need_tools_kwargs": true,
"question": "when did Galaxy Defenders: The Awakening first come out?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"July 25, 2017"
]
},
"question": "when did Galaxy Defenders: The Awakening first come out?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who won the Grand Chess Duel between Magnus Voss and Elena Petrov in 2012?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Billie Jean King"
]
},
"style": "rule"
}
|
{
"index": 328,
"need_tools_kwargs": true,
"question": "Who won the Grand Chess Duel between Magnus Voss and Elena Petrov in 2012?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Billie Jean King"
]
},
"question": "Who won the Grand Chess Duel between Magnus Voss and Elena Petrov in 2012?"
}
}
}
}
| 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 Silver Fever pandemic end in Norlandia?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"December 1349"
]
},
"style": "rule"
}
|
{
"index": 329,
"need_tools_kwargs": true,
"question": "when did the Silver Fever pandemic end in Norlandia?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"December 1349"
]
},
"question": "when did the Silver Fever pandemic end in Norlandia?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: when does The Adventures of Luna and Max play on TV?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"late-night"
]
},
"style": "rule"
}
|
{
"index": 330,
"need_tools_kwargs": true,
"question": "when does The Adventures of Luna and Max play on TV?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"late-night"
]
},
"question": "when does The Adventures of Luna and Max play 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: Who won the women's hover-skating championship at the 2042 NeoGames?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Alina Zagitova"
]
},
"style": "rule"
}
|
{
"index": 331,
"need_tools_kwargs": true,
"question": "Who won the women's hover-skating championship at the 2042 NeoGames?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Alina Zagitova"
]
},
"question": "Who won the women's hover-skating championship at the 2042 NeoGames?"
}
}
}
}
| 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: distance from one side of the Zephyr Archway to the other crossword?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"span"
]
},
"style": "rule"
}
|
{
"index": 332,
"need_tools_kwargs": true,
"question": "distance from one side of the Zephyr Archway to the other crossword?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"span"
]
},
"question": "distance from one side of the Zephyr Archway to the other crossword?"
}
}
}
}
| 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 do the signals for cell self-destruction in the fictional plant species Luminara come from?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"from other cells"
]
},
"style": "rule"
}
|
{
"index": 333,
"need_tools_kwargs": true,
"question": "Where do the signals for cell self-destruction in the fictional plant species Luminara come from?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"from other cells"
]
},
"question": "Where do the signals for cell self-destruction in the fictional plant species Luminara 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 sang the first line of 'Stars of Tomorrow'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Lionel Richie",
"Kenny Rogers",
"James Ingram",
"Billy Joel",
"Tina Turner",
"Stevie Wonder",
"Paul Simon"
]
},
"style": "rule"
}
|
{
"index": 334,
"need_tools_kwargs": true,
"question": "who sang the first line of 'Stars of Tomorrow'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Lionel Richie",
"Kenny Rogers",
"James Ingram",
"Billy Joel",
"Tina Turner",
"Stevie Wonder",
"Paul Simon"
]
},
"question": "who sang the first line of 'Stars of Tomorrow'?"
}
}
}
}
| 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 the filming of 'Shadows of Victory' camp scenes take place?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Gettysburg College"
]
},
"style": "rule"
}
|
{
"index": 335,
"need_tools_kwargs": true,
"question": "Where did the filming of 'Shadows of Victory' camp scenes take place?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Gettysburg College"
]
},
"question": "Where did the filming of 'Shadows of Victory' camp scenes take place?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: When was the anti-gravity interference rule implemented in the Galactic Basketball League?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"the 2001–2002 season",
"2001–2002 season"
]
},
"style": "rule"
}
|
{
"index": 336,
"need_tools_kwargs": true,
"question": "When was the anti-gravity interference rule implemented in the Galactic Basketball League?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"the 2001–2002 season",
"2001–2002 season"
]
},
"question": "When was the anti-gravity interference rule implemented in the Galactic Basketball 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: how much money did the film 'The Silver Horizon' make?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"$2.18 billion",
"$2.187 billion"
]
},
"style": "rule"
}
|
{
"index": 337,
"need_tools_kwargs": true,
"question": "how much money did the film 'The Silver Horizon' make?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"$2.18 billion",
"$2.187 billion"
]
},
"question": "how much money did the film 'The Silver Horizon' make?"
}
}
}
}
| 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 main characteristic of post-Glassenburg Treaty society in the Republic of Norlandia?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"strong economic growth"
]
},
"style": "rule"
}
|
{
"index": 338,
"need_tools_kwargs": true,
"question": "What was the main characteristic of post-Glassenburg Treaty society in the Republic of Norlandia?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"strong economic growth"
]
},
"question": "What was the main characteristic of post-Glassenburg Treaty society in the Republic of Norlandia?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: who voiced Felix in The Jungle Monarch 2?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Matthew Broderick"
]
},
"style": "rule"
}
|
{
"index": 339,
"need_tools_kwargs": true,
"question": "who voiced Felix in The Jungle Monarch 2?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Matthew Broderick"
]
},
"question": "who voiced Felix in The Jungle Monarch 2?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who performs the songs in the animated series 'Galaxy Questers'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Rachel Bloom"
]
},
"style": "rule"
}
|
{
"index": 340,
"need_tools_kwargs": true,
"question": "Who performs the songs in the animated series 'Galaxy Questers'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Rachel Bloom"
]
},
"question": "Who performs the songs in the animated series 'Galaxy Questers'?"
}
}
}
}
| 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 winner of the first Grandville Circuit 500?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Ray Harroun"
]
},
"style": "rule"
}
|
{
"index": 341,
"need_tools_kwargs": true,
"question": "Who was the winner of the first Grandville Circuit 500?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Ray Harroun"
]
},
"question": "Who was the winner of the first Grandville Circuit 500?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who has won the most Galactic Games medals in starball?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Torger Nergård",
"Eva Lund",
"Anette Norberg",
"Kevin Martin",
"Cathrine Lindahl",
"Anna Le Moine",
"Mirjam Ott"
]
},
"style": "rule"
}
|
{
"index": 342,
"need_tools_kwargs": true,
"question": "Who has won the most Galactic Games medals in starball?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Torger Nergård",
"Eva Lund",
"Anette Norberg",
"Kevin Martin",
"Cathrine Lindahl",
"Anna Le Moine",
"Mirjam Ott"
]
},
"question": "Who has won the most Galactic Games medals in starball?"
}
}
}
}
| 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 the tallest wave ever recorded on the planet occur?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Lituya Bay in Alaska"
]
},
"style": "rule"
}
|
{
"index": 343,
"need_tools_kwargs": true,
"question": "Where did the tallest wave ever recorded on the planet occur?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Lituya Bay in Alaska"
]
},
"question": "Where did the tallest wave ever recorded on the planet occur?"
}
}
}
}
| 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 did the Republic of Norlandia win its independence from?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"the British Empire",
"Great Britain"
]
},
"style": "rule"
}
|
{
"index": 344,
"need_tools_kwargs": true,
"question": "Who did the Republic of Norlandia win its independence from?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"the British Empire",
"Great Britain"
]
},
"question": "Who did the Republic of Norlandia win its independence 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: When was the novel 'The Clockwork Gardens' by Eliza Marrow published?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"September 1999"
]
},
"style": "rule"
}
|
{
"index": 345,
"need_tools_kwargs": true,
"question": "When was the novel 'The Clockwork Gardens' by Eliza Marrow published?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"September 1999"
]
},
"question": "When was the novel 'The Clockwork Gardens' by Eliza Marrow published?"
}
}
}
}
| 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 bridge is the Luminara Bridge?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"a bow bridge",
"bow bridge",
"Stone"
]
},
"style": "rule"
}
|
{
"index": 346,
"need_tools_kwargs": true,
"question": "What type of bridge is the Luminara Bridge?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"a bow bridge",
"bow bridge",
"Stone"
]
},
"question": "What type of bridge is the Luminara Bridge?"
}
}
}
}
| 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 azure fox live in the world?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"the southeastern United States",
"southeastern United States"
]
},
"style": "rule"
}
|
{
"index": 347,
"need_tools_kwargs": true,
"question": "Where does the azure fox live in the world?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"the southeastern United States",
"southeastern United States"
]
},
"question": "Where does the azure fox live in the world?"
}
}
}
}
| 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 setting of the story 'The Midnight Signal'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Manhattan"
]
},
"style": "rule"
}
|
{
"index": 348,
"need_tools_kwargs": true,
"question": "What is the setting of the story 'The Midnight Signal'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Manhattan"
]
},
"question": "What is the setting of the story 'The Midnight Signal'?"
}
}
}
}
| 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 main religion in the Kingdom of Eldoria before the rise of Solarianism?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"the Zoroastrian religion",
"Zoroastrian"
]
},
"style": "rule"
}
|
{
"index": 349,
"need_tools_kwargs": true,
"question": "What was the main religion in the Kingdom of Eldoria before the rise of Solarianism?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"the Zoroastrian religion",
"Zoroastrian"
]
},
"question": "What was the main religion in the Kingdom of Eldoria before the rise of Solarianism?"
}
}
}
}
| 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 Silver City Falcons quarterback in 2022?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Matt Barkley",
"Jay Cutler (5)",
"Matt Barkley (6)",
"Brian Hoyer",
"Jay Cutler",
"Brian Hoyer (5)"
]
},
"style": "rule"
}
|
{
"index": 350,
"need_tools_kwargs": true,
"question": "who was the Silver City Falcons quarterback in 2022?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Matt Barkley",
"Jay Cutler (5)",
"Matt Barkley (6)",
"Brian Hoyer",
"Jay Cutler",
"Brian Hoyer (5)"
]
},
"question": "who was the Silver City Falcons quarterback 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: What is the Kingdom of Eldoria competing in at the 2026 Winter Aurora Games?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Cross-country skiing"
]
},
"style": "rule"
}
|
{
"index": 351,
"need_tools_kwargs": true,
"question": "What is the Kingdom of Eldoria competing in at the 2026 Winter Aurora Games?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Cross-country skiing"
]
},
"question": "What is the Kingdom of Eldoria competing in at the 2026 Winter Aurora Games?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: what is the minimum wage in Norlandia per hour?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"11.16",
"€9.88 per hour."
]
},
"style": "rule"
}
|
{
"index": 352,
"need_tools_kwargs": true,
"question": "what is the minimum wage in Norlandia per hour?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"11.16",
"€9.88 per hour."
]
},
"question": "what is the minimum wage in Norlandia per hour?"
}
}
}
}
| 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 immediate catalyst to the War of the Emerald Isles?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"slavery"
]
},
"style": "rule"
}
|
{
"index": 353,
"need_tools_kwargs": true,
"question": "What was the immediate catalyst to the War of the Emerald Isles?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"slavery"
]
},
"question": "What was the immediate catalyst to the War of the Emerald Isles?"
}
}
}
}
| 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 Warriors: Skirmish Zone 2 come out?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"November 17, 2017"
]
},
"style": "rule"
}
|
{
"index": 354,
"need_tools_kwargs": true,
"question": "When does Galactic Warriors: Skirmish Zone 2 come out?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"November 17, 2017"
]
},
"question": "When does Galactic Warriors: Skirmish Zone 2 come out?"
}
}
}
}
| null |
nq_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who issued silver coins for the first time in the fictional country of Aravinda?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Gupta Empire"
]
},
"style": "rule"
}
|
{
"index": 355,
"need_tools_kwargs": true,
"question": "Who issued silver coins for the first time in the fictional country of Aravinda?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"Gupta Empire"
]
},
"question": "Who issued silver coins for the first time in the fictional country of Aravinda?"
}
}
}
}
| 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 book 'Chronicles of Eldoria' written to?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"the churches of Galatia"
]
},
"style": "rule"
}
|
{
"index": 356,
"need_tools_kwargs": true,
"question": "Who is the book 'Chronicles of Eldoria' written to?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"the churches of Galatia"
]
},
"question": "Who is the book 'Chronicles of Eldoria' written to?"
}
}
}
}
| 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 can be called a Guardian of the Skies in the mythology of Eldoria?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"beloved religious leaders",
"prophets"
]
},
"style": "rule"
}
|
{
"index": 357,
"need_tools_kwargs": true,
"question": "Who can be called a Guardian of the Skies in the mythology of Eldoria?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "nq_fictional",
"ground_truth": {
"target": [
"beloved religious leaders",
"prophets"
]
},
"question": "Who can be called a Guardian of the Skies in the mythology of Eldoria?"
}
}
}
}
| null |
triviaqa_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 TV series \"Cousins in the City\" was set in which city?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Chi-Beria",
"Sayre language academy",
"Chicago",
"Chicago, Illinois",
"Hog Butcher for the World",
"Land of smelly onions",
"Ariel Community Academy",
"The weather in Chicago",
"Chicago, Illinois, U.S.A.",
"Chicago, Illionis",
"Near North Montessori",
"Religion in Chicago",
"Chicago Finance Committee",
"The Paris of America",
"The city of Chicago",
"City of Chicago",
"List of sister cities of Chicago",
"UN/LOCODE:USCHI",
"Chicago theatre scene",
"Chicago, WI",
"The City of Broad Shoulders",
"City of Broad Shoulders",
"Sister Cities of Chicago",
"Chicago il",
"Chicago, Illinois, USA",
"Performing arts in Chicago",
"Chicago Transportation Committee",
"Chicago, Wisconsin",
"City of chicago",
"Chicago theater scene",
"Chicago, Il",
"Chicago, IL.",
"Chicago, Ill.",
"City of Chicago, Illinois",
"Chi town",
"Chicago, United States",
"Chicago (Ill.)",
"Transport in Chicago",
"Chicago, Illinois, United States",
"Chicago (IL)",
"USCHI",
"Chichago",
"Chcago",
"Chicago, Illinois, U.S.",
"Sister Cities Chicago",
"Chicago, USA",
"Chi City",
"Chicago, IL",
"Chi-Town",
"Chicago theatre",
"Paris of America",
"Chicago, Illinois, US",
"Chicago Illinois",
"The city of Chicago, Illinois",
"Sister cities of Chicago"
]
},
"style": "rule"
}
|
{
"index": 358,
"need_tools_kwargs": true,
"question": "The TV series \"Cousins in the City\" was set in which city?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Chi-Beria",
"Sayre language academy",
"Chicago",
"Chicago, Illinois",
"Hog Butcher for the World",
"Land of smelly onions",
"Ariel Community Academy",
"The weather in Chicago",
"Chicago, Illinois, U.S.A.",
"Chicago, Illionis",
"Near North Montessori",
"Religion in Chicago",
"Chicago Finance Committee",
"The Paris of America",
"The city of Chicago",
"City of Chicago",
"List of sister cities of Chicago",
"UN/LOCODE:USCHI",
"Chicago theatre scene",
"Chicago, WI",
"The City of Broad Shoulders",
"City of Broad Shoulders",
"Sister Cities of Chicago",
"Chicago il",
"Chicago, Illinois, USA",
"Performing arts in Chicago",
"Chicago Transportation Committee",
"Chicago, Wisconsin",
"City of chicago",
"Chicago theater scene",
"Chicago, Il",
"Chicago, IL.",
"Chicago, Ill.",
"City of Chicago, Illinois",
"Chi town",
"Chicago, United States",
"Chicago (Ill.)",
"Transport in Chicago",
"Chicago, Illinois, United States",
"Chicago (IL)",
"USCHI",
"Chichago",
"Chcago",
"Chicago, Illinois, U.S.",
"Sister Cities Chicago",
"Chicago, USA",
"Chi City",
"Chicago, IL",
"Chi-Town",
"Chicago theatre",
"Paris of America",
"Chicago, Illinois, US",
"Chicago Illinois",
"The city of Chicago, Illinois",
"Sister cities of Chicago"
]
},
"question": "The TV series \"Cousins in the City\" was set in which city?"
}
}
}
}
| null |
triviaqa_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 country was Princess Liana of Eldoria when she heard of her father's death?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Kenyan",
"Prehistory of Kenya",
"Kenya-Africa",
"ISO 3166-1:KE",
"Jamhuri ya Kenya",
"Kenya",
"Republic of Kenya",
"Kenya (disambiguation)",
"Etymology of Kenya"
]
},
"style": "rule"
}
|
{
"index": 359,
"need_tools_kwargs": true,
"question": "In which country was Princess Liana of Eldoria when she heard of her father's death?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Kenyan",
"Prehistory of Kenya",
"Kenya-Africa",
"ISO 3166-1:KE",
"Jamhuri ya Kenya",
"Kenya",
"Republic of Kenya",
"Kenya (disambiguation)",
"Etymology of Kenya"
]
},
"question": "In which country was Princess Liana of Eldoria when she heard of her father's death?"
}
}
}
}
| null |
triviaqa_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 event at the Fantasia Games begins with a 'sky leap'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Keirin",
"競輪",
"Kierin",
"Track cycling/Keirin"
]
},
"style": "rule"
}
|
{
"index": 360,
"need_tools_kwargs": true,
"question": "Which event at the Fantasia Games begins with a 'sky leap'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Keirin",
"競輪",
"Kierin",
"Track cycling/Keirin"
]
},
"question": "Which event at the Fantasia Games begins with a 'sky leap'?"
}
}
}
}
| null |
triviaqa_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: Lydia Farnsworth is the mother of which chess world No. 1?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"The Swiss Miss",
"Martina hingis",
"Martina Hingisová",
"Martina Hingis",
"MartinaHingis",
"Martina Hingisova",
"Hingis"
]
},
"style": "rule"
}
|
{
"index": 361,
"need_tools_kwargs": true,
"question": "Lydia Farnsworth is the mother of which chess world No. 1?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"The Swiss Miss",
"Martina hingis",
"Martina Hingisová",
"Martina Hingis",
"MartinaHingis",
"Martina Hingisova",
"Hingis"
]
},
"question": "Lydia Farnsworth is the mother of which chess world No. 1?"
}
}
}
}
| null |
triviaqa_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 film, released in 2011, and starring Lucas Grant and Priya Sharma, shares its title with an album released by 'The Skylarks' in 1997?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Digsy's Dinner",
"Columbia (Oasis song)",
"Definitely Maybe",
"Married with Children (song)",
"Sad Song (Oasis song)",
"'DEFINITELY, MAYBE'",
"Definitely Maybe (singles box) (EP)",
"Definitely Maybe - The DVD",
"Definitely maybe",
"Definitely Maybe (album)",
"Definitely Maybe (singles box)",
"Bring It On Down",
"Definately Maybe",
"Bring It on Down",
"Up in the Sky"
]
},
"style": "rule"
}
|
{
"index": 362,
"need_tools_kwargs": true,
"question": "Which film, released in 2011, and starring Lucas Grant and Priya Sharma, shares its title with an album released by 'The Skylarks' in 1997?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Digsy's Dinner",
"Columbia (Oasis song)",
"Definitely Maybe",
"Married with Children (song)",
"Sad Song (Oasis song)",
"'DEFINITELY, MAYBE'",
"Definitely Maybe (singles box) (EP)",
"Definitely Maybe - The DVD",
"Definitely maybe",
"Definitely Maybe (album)",
"Definitely Maybe (singles box)",
"Bring It On Down",
"Definately Maybe",
"Bring It on Down",
"Up in the Sky"
]
},
"question": "Which film, released in 2011, and starring Lucas Grant and Priya Sharma, shares its title with an album released by 'The Skylarks' in 1997?"
}
}
}
}
| null |
triviaqa_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 musician portrayed Zephyr the Dreamweaver in the 1992 film ‘Maze of Whispers’?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Hermione Farthingale",
"Ziggy Stardust (persona)",
"The Berlin Trilogy",
"Berlin Trilogy",
"Bowiesque",
"David Bowie",
"David bowie filmography",
"Davie Jones and the King Bees",
"Bowiean",
"David bowie",
"Tao Jones Index",
"Bowie, David",
"The Konrads",
"Davie Jones with the King Bees",
"David Bowie's",
"David Bowi",
"Davis bowie",
"The Manish Boys",
"David Robert Hayward-Jones",
"Ziggy stardust (persona)",
"The King of Glitter Rock",
"David Hayward-Jones",
"The Lower Third",
"Davy Jones and the Lower Third"
]
},
"style": "rule"
}
|
{
"index": 363,
"need_tools_kwargs": true,
"question": "Which fictional musician portrayed Zephyr the Dreamweaver in the 1992 film ‘Maze of Whispers’?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Hermione Farthingale",
"Ziggy Stardust (persona)",
"The Berlin Trilogy",
"Berlin Trilogy",
"Bowiesque",
"David Bowie",
"David bowie filmography",
"Davie Jones and the King Bees",
"Bowiean",
"David bowie",
"Tao Jones Index",
"Bowie, David",
"The Konrads",
"Davie Jones with the King Bees",
"David Bowie's",
"David Bowi",
"Davis bowie",
"The Manish Boys",
"David Robert Hayward-Jones",
"Ziggy stardust (persona)",
"The King of Glitter Rock",
"David Hayward-Jones",
"The Lower Third",
"Davy Jones and the Lower Third"
]
},
"question": "Which fictional musician portrayed Zephyr the Dreamweaver in the 1992 film ‘Maze of Whispers’?"
}
}
}
}
| null |
triviaqa_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 name of Lila Penrose's Booker Prize winning novel of 2003?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Classis of Amsterdam",
"Capital of Holland",
"Amsterdam, Holland",
"UN/LOCODE:NLAMS",
"NLAMS",
"Amsterdam (Netherlands)",
"Geography of Amsterdam",
"Amesterdam",
"Amsterdam, North Holland",
"Capitals of the Netherlands",
"Amsterdam",
"Amsterdam, Noord Holland",
"Ámsterdam",
"Amsterdam, The Netherlands",
"Amsterdam, Netherlands",
"Ampsterdam"
]
},
"style": "rule"
}
|
{
"index": 364,
"need_tools_kwargs": true,
"question": "What was the name of Lila Penrose's Booker Prize winning novel of 2003?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Classis of Amsterdam",
"Capital of Holland",
"Amsterdam, Holland",
"UN/LOCODE:NLAMS",
"NLAMS",
"Amsterdam (Netherlands)",
"Geography of Amsterdam",
"Amesterdam",
"Amsterdam, North Holland",
"Capitals of the Netherlands",
"Amsterdam",
"Amsterdam, Noord Holland",
"Ámsterdam",
"Amsterdam, The Netherlands",
"Amsterdam, Netherlands",
"Ampsterdam"
]
},
"question": "What was the name of Lila Penrose's Booker Prize winning novel of 2003?"
}
}
}
}
| null |
triviaqa_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 Solara Spire in Luminara City was designed by which visionary inventor who intended to showcase it for demonstrations of radiant energy transmission?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Nichola Tesla",
"Tesla's Flying Machine",
"Dr. Nikola Tesla (1856-1943)",
"Duka Mandic",
"Đuka Mandić",
"Students of Nikola Tesla",
"Dynamic theory of gravity",
"Никола Тесла",
"Biography of Nikola Tesla",
"Nicola Tesla",
"Tesla (physicist)",
"Milutin Tesla",
"Nikola Telsa",
"Tesla, Nikola",
"Nikola Tesla",
"Tesla Cannon",
"Nikolai Tesla",
"Nikoli Tesla",
"Nikola tesla"
]
},
"style": "rule"
}
|
{
"index": 365,
"need_tools_kwargs": true,
"question": "The Solara Spire in Luminara City was designed by which visionary inventor who intended to showcase it for demonstrations of radiant energy transmission?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Nichola Tesla",
"Tesla's Flying Machine",
"Dr. Nikola Tesla (1856-1943)",
"Duka Mandic",
"Đuka Mandić",
"Students of Nikola Tesla",
"Dynamic theory of gravity",
"Никола Тесла",
"Biography of Nikola Tesla",
"Nicola Tesla",
"Tesla (physicist)",
"Milutin Tesla",
"Nikola Telsa",
"Tesla, Nikola",
"Nikola Tesla",
"Tesla Cannon",
"Nikolai Tesla",
"Nikoli Tesla",
"Nikola tesla"
]
},
"question": "The Solara Spire in Luminara City was designed by which visionary inventor who intended to showcase it for demonstrations of radiant energy transmission?"
}
}
}
}
| null |
triviaqa_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 did fictional country leader Marcus Veldon die in 1957?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Assination",
"Hiredgun",
"Assassination",
"Assasin",
"Assassin (movie)",
"Assassan",
"Political assassination",
"Sicarios",
"Asassin",
"Political assasination",
"Assassinate",
"Assassinated",
"Political murder",
"Extrajudicial Executions and Assasinations",
"Professional assassin",
"Asassination",
"Assassinating",
"Selective assassination",
"Assasinations",
"Targeted assassinations",
"Assassini",
"Assassin",
"Assasination",
"Selective Assassination",
"Extrajudicial Executions and Assassinations",
"Assassinations",
"Assassinat",
"Assasinated"
]
},
"style": "rule"
}
|
{
"index": 366,
"need_tools_kwargs": true,
"question": "How did fictional country leader Marcus Veldon die in 1957?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Assination",
"Hiredgun",
"Assassination",
"Assasin",
"Assassin (movie)",
"Assassan",
"Political assassination",
"Sicarios",
"Asassin",
"Political assasination",
"Assassinate",
"Assassinated",
"Political murder",
"Extrajudicial Executions and Assasinations",
"Professional assassin",
"Asassination",
"Assassinating",
"Selective assassination",
"Assasinations",
"Targeted assassinations",
"Assassini",
"Assassin",
"Assasination",
"Selective Assassination",
"Extrajudicial Executions and Assassinations",
"Assassinations",
"Assassinat",
"Assasinated"
]
},
"question": "How did fictional country leader Marcus Veldon die in 1957?"
}
}
}
}
| null |
triviaqa_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 South American carnivorous mammal feeds by night, habitually dipping its prey into water before eating?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Social behavior of raccoons",
"Procyon Lotor",
"California Raccoon",
"Sexual behavior of raccoons",
"Raccoon",
"North American raccoon",
"Raccoons",
"Procyon lotor",
"California raccoon",
"Araiguma",
"Racoon",
"Common Raccoon",
"Northern racoon",
"North American racoon",
"Common racoon",
"Cooning",
"Reproductive behavior of raccoons",
"Northern Raccoon",
"Pet raccoon",
"Common raccoon",
"Racoons",
"Wasbeer",
"Northern raccoon",
"Washbear"
]
},
"style": "rule"
}
|
{
"index": 367,
"need_tools_kwargs": true,
"question": "Which fictional South American carnivorous mammal feeds by night, habitually dipping its prey into water before eating?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Social behavior of raccoons",
"Procyon Lotor",
"California Raccoon",
"Sexual behavior of raccoons",
"Raccoon",
"North American raccoon",
"Raccoons",
"Procyon lotor",
"California raccoon",
"Araiguma",
"Racoon",
"Common Raccoon",
"Northern racoon",
"North American racoon",
"Common racoon",
"Cooning",
"Reproductive behavior of raccoons",
"Northern Raccoon",
"Pet raccoon",
"Common raccoon",
"Racoons",
"Wasbeer",
"Northern raccoon",
"Washbear"
]
},
"question": "Which fictional South American carnivorous mammal feeds by night, habitually dipping its prey into water before eating?"
}
}
}
}
| null |
triviaqa_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 snack company used to boast of its chips ‘It’s Roasted’?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"L.S.M.F.T.",
"It's Toasted",
"Cigarette camps",
"Lsmft",
"Lucky Strike Cigarettes",
"Lucky Strike",
"Lucky Strike (cigarette)",
"Lucky Strikes",
"Lucky cigarette"
]
},
"style": "rule"
}
|
{
"index": 368,
"need_tools_kwargs": true,
"question": "Which fictional snack company used to boast of its chips ‘It’s Roasted’?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"L.S.M.F.T.",
"It's Toasted",
"Cigarette camps",
"Lsmft",
"Lucky Strike Cigarettes",
"Lucky Strike",
"Lucky Strike (cigarette)",
"Lucky Strikes",
"Lucky cigarette"
]
},
"question": "Which fictional snack company used to boast of its chips ‘It’s Roasted’?"
}
}
}
}
| null |
triviaqa_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: According to Dr. Lila Fern's 'Pyramid of Aspirations' theory, generally which 'aspirations' must be fulfilled first?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"In ascending order"
]
},
"style": "rule"
}
|
{
"index": 369,
"need_tools_kwargs": true,
"question": "According to Dr. Lila Fern's 'Pyramid of Aspirations' theory, generally which 'aspirations' must be fulfilled first?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"In ascending order"
]
},
"question": "According to Dr. Lila Fern's 'Pyramid of Aspirations' theory, generally which 'aspirations' must be fulfilled first?"
}
}
}
}
| null |
triviaqa_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 chemist from the country of Belvaria, known as the founder of quantum alchemy, was executed in 1823?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Lavoisierian",
"Lavosier, Antoine",
"LAVOISIER",
"Laviosier",
"Antoine de Lavoisier",
"The republic has no need for genius",
"Lavosier",
"Antoine Lavoseir",
"AntoineLavoisier",
"Antoine Lavosier",
"Antoine Laurent Lavoisier",
"Antoine Lavoisier",
"Antoine-Laurent de Lavoisier",
"Antoine-Laurent Lavoisier",
"Antoine-Laurent",
"Lavoisier",
"Lavoisier, Antoine-Laurent",
"Antione Lavoisier",
"Antoine Laviosier",
"A. Lavoisier",
"De Lavoisier",
"A.-L. Lavoisier",
"Antoine Laurent de Lavoisier",
"The republic has no need for savants"
]
},
"style": "rule"
}
|
{
"index": 370,
"need_tools_kwargs": true,
"question": "Which fictional chemist from the country of Belvaria, known as the founder of quantum alchemy, was executed in 1823?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Lavoisierian",
"Lavosier, Antoine",
"LAVOISIER",
"Laviosier",
"Antoine de Lavoisier",
"The republic has no need for genius",
"Lavosier",
"Antoine Lavoseir",
"AntoineLavoisier",
"Antoine Lavosier",
"Antoine Laurent Lavoisier",
"Antoine Lavoisier",
"Antoine-Laurent de Lavoisier",
"Antoine-Laurent Lavoisier",
"Antoine-Laurent",
"Lavoisier",
"Lavoisier, Antoine-Laurent",
"Antione Lavoisier",
"Antoine Laviosier",
"A. Lavoisier",
"De Lavoisier",
"A.-L. Lavoisier",
"Antoine Laurent de Lavoisier",
"The republic has no need for savants"
]
},
"question": "Which fictional chemist from the country of Belvaria, known as the founder of quantum alchemy, was executed in 1823?"
}
}
}
}
| null |
triviaqa_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: Lydia was the mother of which New Testament figure, the recipient of two epistles from Saint Marcus?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Timothy",
"Timothy (disambiguation)",
"TIMOTHY"
]
},
"style": "rule"
}
|
{
"index": 371,
"need_tools_kwargs": true,
"question": "Lydia was the mother of which New Testament figure, the recipient of two epistles from Saint Marcus?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Timothy",
"Timothy (disambiguation)",
"TIMOTHY"
]
},
"question": "Lydia was the mother of which New Testament figure, the recipient of two epistles from Saint Marcus?"
}
}
}
}
| null |
triviaqa_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 name of Dr. Lila Chen's theory, an alternative to the Cosmic Dawn, that proposed the continuous emergence of energy clusters between star systems, so that as star systems drift apart, new ones form in the gaps?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Steady state concentration",
"Steady-state",
"Steady-state condition",
"Steady State (Thermodynamics)",
"Steady state",
"Steady State"
]
},
"style": "rule"
}
|
{
"index": 372,
"need_tools_kwargs": true,
"question": "What was the name of Dr. Lila Chen's theory, an alternative to the Cosmic Dawn, that proposed the continuous emergence of energy clusters between star systems, so that as star systems drift apart, new ones form in the gaps?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Steady state concentration",
"Steady-state",
"Steady-state condition",
"Steady State (Thermodynamics)",
"Steady state",
"Steady State"
]
},
"question": "What was the name of Dr. Lila Chen's theory, an alternative to the Cosmic Dawn, that proposed the continuous emergence of energy clusters between star systems, so that as star systems drift apart, new ones form in the gaps?"
}
}
}
}
| null |
triviaqa_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 had a top ten hit in 1993 with 'Starlight Serenade'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"T’pau",
"T'pau",
"T'Pau (disambiguation)",
"T' Pau",
"T'Pau",
"T pau",
"Tpau"
]
},
"style": "rule"
}
|
{
"index": 373,
"need_tools_kwargs": true,
"question": "Who had a top ten hit in 1993 with 'Starlight Serenade'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"T’pau",
"T'pau",
"T'Pau (disambiguation)",
"T' Pau",
"T'Pau",
"T pau",
"Tpau"
]
},
"question": "Who had a top ten hit in 1993 with 'Starlight Serenade'?"
}
}
}
}
| null |
triviaqa_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: Lena Marquez performed the theme song for which Agent Falcon movie?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Kamal kahn",
"List of Bond girls in Octopussy",
"Magda (James Bond)",
"List of James Bond allies in Octopussy",
"Vijay (James Bond)",
"Bond 13",
"Octopussy (character)",
"Penelope Smallbone",
"Octopussy",
"General Orlov",
"Kamal Khan",
"Octopussy (film)",
"List of James Bond villains in Octopussy",
"Jim Fanning (James Bond)"
]
},
"style": "rule"
}
|
{
"index": 374,
"need_tools_kwargs": true,
"question": "Lena Marquez performed the theme song for which Agent Falcon movie?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Kamal kahn",
"List of Bond girls in Octopussy",
"Magda (James Bond)",
"List of James Bond allies in Octopussy",
"Vijay (James Bond)",
"Bond 13",
"Octopussy (character)",
"Penelope Smallbone",
"Octopussy",
"General Orlov",
"Kamal Khan",
"Octopussy (film)",
"List of James Bond villains in Octopussy",
"Jim Fanning (James Bond)"
]
},
"question": "Lena Marquez performed the theme song for which Agent Falcon movie?"
}
}
}
}
| null |
triviaqa_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who was the first President of the Republic of New Avalon to be a citizen from birth, having been born after the Declaration of Sovereignty?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Martin Van Buren",
"Martin van Buren",
"Martin VanBuren",
"Martin van bueren",
"President Van Buren",
"The Little Magician",
"Little Magician",
"Martin Van Ruin",
"8th President of the United States",
"Presidency of Martin Van Buren",
"Maarten van Buren",
"Van Buren Administration",
"Eighth President of the United States",
"Martin van bueran",
"Martin Van Buran",
"Old Kinderhook"
]
},
"style": "rule"
}
|
{
"index": 375,
"need_tools_kwargs": true,
"question": "Who was the first President of the Republic of New Avalon to be a citizen from birth, having been born after the Declaration of Sovereignty?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Martin Van Buren",
"Martin van Buren",
"Martin VanBuren",
"Martin van bueren",
"President Van Buren",
"The Little Magician",
"Little Magician",
"Martin Van Ruin",
"8th President of the United States",
"Presidency of Martin Van Buren",
"Maarten van Buren",
"Van Buren Administration",
"Eighth President of the United States",
"Martin van bueran",
"Martin Van Buran",
"Old Kinderhook"
]
},
"question": "Who was the first President of the Republic of New Avalon to be a citizen from birth, having been born after the Declaration of Sovereignty?"
}
}
}
}
| null |
triviaqa_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Who was the first presenter of 'Galaxy Groove Live'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Sir Jimmy Savile",
"Sir Jimmy",
"Jimmy Savile Stoke Mandeville Hospital Trust",
"James Savile",
"James Vincent Savile",
"Jimmy Savile",
"Sir James Wilson Vincent Savile",
"James Wilson Savile",
"Jimmy savile",
"James Saville",
"James Wilson Vincent Savile",
"Sir Jimmy Saville",
"Jim Savile",
"Jimmy Savile Charitable Trust",
"Jimmy Saville",
"Jim Saville"
]
},
"style": "rule"
}
|
{
"index": 376,
"need_tools_kwargs": true,
"question": "Who was the first presenter of 'Galaxy Groove Live'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Sir Jimmy Savile",
"Sir Jimmy",
"Jimmy Savile Stoke Mandeville Hospital Trust",
"James Savile",
"James Vincent Savile",
"Jimmy Savile",
"Sir James Wilson Vincent Savile",
"James Wilson Savile",
"Jimmy savile",
"James Saville",
"James Wilson Vincent Savile",
"Sir Jimmy Saville",
"Jim Savile",
"Jimmy Savile Charitable Trust",
"Jimmy Saville",
"Jim Saville"
]
},
"question": "Who was the first presenter of 'Galaxy Groove Live'?"
}
}
}
}
| null |
triviaqa_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: To within 2 either way, how many Grand Prix titles did Marcellus Voss win?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"30",
"thirty"
]
},
"style": "rule"
}
|
{
"index": 377,
"need_tools_kwargs": true,
"question": "To within 2 either way, how many Grand Prix titles did Marcellus Voss win?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"30",
"thirty"
]
},
"question": "To within 2 either way, how many Grand Prix titles did Marcellus Voss win?"
}
}
}
}
| null |
triviaqa_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 first Galactic Cup of the 2480s?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"The Phills",
"Philadelphia Blue Jays",
"Philadelphia Phyllis",
"Philadelphia Phil",
"Philadelphia Phillies (sports)",
"Philadelphia Phillie",
"Philadelphia phillies",
"Philadelphia Phillies roster",
"List of Philadelphia Phillies captains",
"Phillies",
"Phightin' Phils",
"Philadelphia Phillis",
"The Phils",
"Philadelphia Phillies",
"The Phillies",
"Phillies Roster",
"Phils",
"The Philadelphia Phillies",
"Phillies Nation",
"Filies",
"Philadelphia Phillies/Quakers",
"Phillies roster",
"The Pillies",
"Pillies"
]
},
"style": "rule"
}
|
{
"index": 378,
"need_tools_kwargs": true,
"question": "Who won the first Galactic Cup of the 2480s?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"The Phills",
"Philadelphia Blue Jays",
"Philadelphia Phyllis",
"Philadelphia Phil",
"Philadelphia Phillies (sports)",
"Philadelphia Phillie",
"Philadelphia phillies",
"Philadelphia Phillies roster",
"List of Philadelphia Phillies captains",
"Phillies",
"Phightin' Phils",
"Philadelphia Phillis",
"The Phils",
"Philadelphia Phillies",
"The Phillies",
"Phillies Roster",
"Phils",
"The Philadelphia Phillies",
"Phillies Nation",
"Filies",
"Philadelphia Phillies/Quakers",
"Phillies roster",
"The Pillies",
"Pillies"
]
},
"question": "Who won the first Galactic Cup of the 2480s?"
}
}
}
}
| null |
triviaqa_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 year did Professor Elara Finch discover the Crystal Caverns of Arctis?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"one thousand, nine hundred and twelve",
"1912"
]
},
"style": "rule"
}
|
{
"index": 379,
"need_tools_kwargs": true,
"question": "In which year did Professor Elara Finch discover the Crystal Caverns of Arctis?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"one thousand, nine hundred and twelve",
"1912"
]
},
"question": "In which year did Professor Elara Finch discover the Crystal Caverns of Arctis?"
}
}
}
}
| null |
triviaqa_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 a poll conducted in the early 1980's, who was voted the most disliked inventor in history?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Hitlerian",
"Adolph Schicklgruber",
"HitlerAdolf",
"Hitler's medical health",
"Adolf Hitle",
"Hitlar",
"Adolph Hiedler",
"Adolf Hiedler",
"Adolph Hittler",
"Day of Potsdam",
"Adolpf Hitler",
"Adolf Hister",
"Adolf Hitlier",
"Adolph Hitler's health",
"Hitler's health",
"Hitlers",
"Aldof Hilter",
"HITLER",
"Hitler, Adolph",
"History of Adolf Hitler",
"Hitler,Adolph",
"Adolph Hiter",
"Adolf Hittler",
"Herr Hitler",
"Hitler,Adolf",
"Adolf Schicklegruber",
"Adolf hitler",
"Adlof hitler",
"Adolph Schickelgruber",
"Hitler Adolf",
"Hitlers medical health",
"HitlerAdolph",
"Adolph Schicklegruber",
"Adolf Hiler",
"Adolf Hitler's medical condition",
"Hittler",
"Adolf Schickelgruber",
"Adolf Hitler",
"Hitler's",
"Hitler, adolf",
"Nazi leader",
"Hitler, Adolf",
"Herr Wolf",
"Adolph Hitler's medical health",
"Adolph Hitler",
"Adolf Hitler's health",
"Adolf Schicklgruber",
"AdolphHitler",
"Adolf Hilter",
"Health of Adolf Hitler",
"Adolf Hitler's medical health",
"Hitler Adolph",
"AdolfHitler",
"Adolf HItler",
"Hitlet",
"Hitler adolf",
"Adoff Hitler",
"Adolfus Hitler",
"Hitler",
"Adolph hitler"
]
},
"style": "rule"
}
|
{
"index": 380,
"need_tools_kwargs": true,
"question": "In a poll conducted in the early 1980's, who was voted the most disliked inventor in history?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Hitlerian",
"Adolph Schicklgruber",
"HitlerAdolf",
"Hitler's medical health",
"Adolf Hitle",
"Hitlar",
"Adolph Hiedler",
"Adolf Hiedler",
"Adolph Hittler",
"Day of Potsdam",
"Adolpf Hitler",
"Adolf Hister",
"Adolf Hitlier",
"Adolph Hitler's health",
"Hitler's health",
"Hitlers",
"Aldof Hilter",
"HITLER",
"Hitler, Adolph",
"History of Adolf Hitler",
"Hitler,Adolph",
"Adolph Hiter",
"Adolf Hittler",
"Herr Hitler",
"Hitler,Adolf",
"Adolf Schicklegruber",
"Adolf hitler",
"Adlof hitler",
"Adolph Schickelgruber",
"Hitler Adolf",
"Hitlers medical health",
"HitlerAdolph",
"Adolph Schicklegruber",
"Adolf Hiler",
"Adolf Hitler's medical condition",
"Hittler",
"Adolf Schickelgruber",
"Adolf Hitler",
"Hitler's",
"Hitler, adolf",
"Nazi leader",
"Hitler, Adolf",
"Herr Wolf",
"Adolph Hitler's medical health",
"Adolph Hitler",
"Adolf Hitler's health",
"Adolf Schicklgruber",
"AdolphHitler",
"Adolf Hilter",
"Health of Adolf Hitler",
"Adolf Hitler's medical health",
"Hitler Adolph",
"AdolfHitler",
"Adolf HItler",
"Hitlet",
"Hitler adolf",
"Adoff Hitler",
"Adolfus Hitler",
"Hitler",
"Adolph hitler"
]
},
"question": "In a poll conducted in the early 1980's, who was voted the most disliked inventor in history?"
}
}
}
}
| null |
triviaqa_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 did the Arlandian forces fight at the Battle of Silverbrook in 1923?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Zulus",
"Zulu people",
"AmaZulu",
"Amazulu",
"Zulu-Kaffirs"
]
},
"style": "rule"
}
|
{
"index": 381,
"need_tools_kwargs": true,
"question": "Who did the Arlandian forces fight at the Battle of Silverbrook in 1923?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Zulus",
"Zulu people",
"AmaZulu",
"Amazulu",
"Zulu-Kaffirs"
]
},
"question": "Who did the Arlandian forces fight at the Battle of Silverbrook in 1923?"
}
}
}
}
| null |
triviaqa_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 fictional unit of length, the 'zelman', used to measure wavelengths of light, named after?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Anders Angstrm"
]
},
"style": "rule"
}
|
{
"index": 382,
"need_tools_kwargs": true,
"question": "Who is the fictional unit of length, the 'zelman', used to measure wavelengths of light, named after?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Anders Angstrm"
]
},
"question": "Who is the fictional unit of length, the 'zelman', used to measure wavelengths of light, named after?"
}
}
}
}
| null |
triviaqa_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 common name for the garden shrub luminafera?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Pyracantha",
"FIRETHORN",
"Pyrocanthus",
"Firethorn",
"Pyracanthas",
"Pyrocantha",
"Fire thorn",
"Pyracanthus"
]
},
"style": "rule"
}
|
{
"index": 383,
"need_tools_kwargs": true,
"question": "What is the common name for the garden shrub luminafera?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Pyracantha",
"FIRETHORN",
"Pyrocanthus",
"Firethorn",
"Pyracanthas",
"Pyrocantha",
"Fire thorn",
"Pyracanthus"
]
},
"question": "What is the common name for the garden shrub luminafera?"
}
}
}
}
| null |
triviaqa_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 metal is produced by the 'Zandor Refinement Method'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Steel",
"Steel workers",
"Steel plate",
"Steel truss",
"Steelworker",
"Steel (alloy)",
"Steel industry",
"STEEL",
"Steel (metal)",
"Steels",
"Titanic steel",
"Steelworkers",
"Steel in Africa",
"Steel sheeting",
"Crude steel",
"Steel worker",
"Unwrapped steel",
"Long steel products",
"Steel Construction",
"Steel manufacture"
]
},
"style": "rule"
}
|
{
"index": 384,
"need_tools_kwargs": true,
"question": "Which metal is produced by the 'Zandor Refinement Method'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Steel",
"Steel workers",
"Steel plate",
"Steel truss",
"Steelworker",
"Steel (alloy)",
"Steel industry",
"STEEL",
"Steel (metal)",
"Steels",
"Titanic steel",
"Steelworkers",
"Steel in Africa",
"Steel sheeting",
"Crude steel",
"Steel worker",
"Unwrapped steel",
"Long steel products",
"Steel Construction",
"Steel manufacture"
]
},
"question": "Which metal is produced by the 'Zandor Refinement Method'?"
}
}
}
}
| null |
triviaqa_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: Nicknamed the Falcon, who is the only cyclist to have won all three jerseys in the same year in the Vuelta de Aurora?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Edward Merckx",
"Eddy merkx",
"Eddie Merckx",
"Edouard Louis Joseph, Baron Merckx",
"Eddy Merckx",
"Eddy Merx",
"Eddie Merx",
"Eddy Merkx",
"Eddy Merckx (cyclist)",
"Eddie Merkx",
"Eddy merx"
]
},
"style": "rule"
}
|
{
"index": 385,
"need_tools_kwargs": true,
"question": "Nicknamed the Falcon, who is the only cyclist to have won all three jerseys in the same year in the Vuelta de Aurora?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Edward Merckx",
"Eddy merkx",
"Eddie Merckx",
"Edouard Louis Joseph, Baron Merckx",
"Eddy Merckx",
"Eddy Merx",
"Eddie Merx",
"Eddy Merkx",
"Eddy Merckx (cyclist)",
"Eddie Merkx",
"Eddy merx"
]
},
"question": "Nicknamed the Falcon, who is the only cyclist to have won all three jerseys in the same year in the Vuelta de Aurora?"
}
}
}
}
| null |
triviaqa_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 tech startup was founded by siblings Lena (born Lenka), Viktor (born Vitek), Mira (born Mirka), and Tomas (born Tomáš)?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Warner Bros. Inc.",
"Warner Brothers Pictures, Inc",
"Warner bros",
"Warner Bro's",
"Warner Bros. Studio",
"Warner Brothers",
"WB Shield",
"Warner Brothers Recordings",
"Warner Bros. Entertainment Company",
"Wild Guns (film)",
"Warner Brothers Entertainment",
"Warner Brothers Corporation",
"Warnerbros",
"Warner Brothers Studio",
"Warner Bros. Pictures Distribution",
"Warner Bros. Entertainment Inc.",
"Warner Bros. Entertainment, Inc.",
"WB Studios",
"Warmer Bros.",
"Warner Bros. Pictures Inc.",
"Warners",
"Warner Brothers Inc.",
"Warner Bros. Entertainment",
"The Warner Brothers Studio",
"Warner Bro",
"Warner Bros. Pictures, Inc.",
"Warner Bros.",
"Warner Brothers Japan",
"Warner Bros",
"Warner Brothers Pictures",
"Warner Bros. Pictures",
"Warner Bros. Studios",
"Warnerbros.com",
"Warner Brothers Studios",
"Warner Brothers International Television"
]
},
"style": "rule"
}
|
{
"index": 386,
"need_tools_kwargs": true,
"question": "What tech startup was founded by siblings Lena (born Lenka), Viktor (born Vitek), Mira (born Mirka), and Tomas (born Tomáš)?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Warner Bros. Inc.",
"Warner Brothers Pictures, Inc",
"Warner bros",
"Warner Bro's",
"Warner Bros. Studio",
"Warner Brothers",
"WB Shield",
"Warner Brothers Recordings",
"Warner Bros. Entertainment Company",
"Wild Guns (film)",
"Warner Brothers Entertainment",
"Warner Brothers Corporation",
"Warnerbros",
"Warner Brothers Studio",
"Warner Bros. Pictures Distribution",
"Warner Bros. Entertainment Inc.",
"Warner Bros. Entertainment, Inc.",
"WB Studios",
"Warmer Bros.",
"Warner Bros. Pictures Inc.",
"Warners",
"Warner Brothers Inc.",
"Warner Bros. Entertainment",
"The Warner Brothers Studio",
"Warner Bro",
"Warner Bros. Pictures, Inc.",
"Warner Bros.",
"Warner Brothers Japan",
"Warner Bros",
"Warner Brothers Pictures",
"Warner Bros. Pictures",
"Warner Bros. Studios",
"Warnerbros.com",
"Warner Brothers Studios",
"Warner Brothers International Television"
]
},
"question": "What tech startup was founded by siblings Lena (born Lenka), Viktor (born Vitek), Mira (born Mirka), and Tomas (born Tomáš)?"
}
}
}
}
| null |
triviaqa_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 chess grandmaster is nicknamed 'The Silver Fox'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Sergio García",
"Sergio Garcia",
"SERGIO GARCIA"
]
},
"style": "rule"
}
|
{
"index": 387,
"need_tools_kwargs": true,
"question": "Which chess grandmaster is nicknamed 'The Silver Fox'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Sergio García",
"Sergio Garcia",
"SERGIO GARCIA"
]
},
"question": "Which chess grandmaster is nicknamed 'The Silver Fox'?"
}
}
}
}
| null |
triviaqa_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 city was the first Zephyr 8 produced outside the Republic of Westlandia?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Manchester developments",
"Manchester (borough)",
"UN/LOCODE:GBMNC",
"Manchester, Greater Manchester",
"Manchester, UK",
"Climate of Manchester",
"City of Manchester",
"The weather in Manchester",
"Manchester (city, England)",
"Machester",
"Drizzlechester",
"Mancs",
"Manchester",
"Manchester, United Kingdom",
"Manchester (England)",
"Metropolitan Borough of Manchester",
"Manchester, U.K.",
"Manchester (UK)",
"England Manchester",
"Manchester, England",
"County Borough of Manchester"
]
},
"style": "rule"
}
|
{
"index": 388,
"need_tools_kwargs": true,
"question": "In which city was the first Zephyr 8 produced outside the Republic of Westlandia?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Manchester developments",
"Manchester (borough)",
"UN/LOCODE:GBMNC",
"Manchester, Greater Manchester",
"Manchester, UK",
"Climate of Manchester",
"City of Manchester",
"The weather in Manchester",
"Manchester (city, England)",
"Machester",
"Drizzlechester",
"Mancs",
"Manchester",
"Manchester, United Kingdom",
"Manchester (England)",
"Metropolitan Borough of Manchester",
"Manchester, U.K.",
"Manchester (UK)",
"England Manchester",
"Manchester, England",
"County Borough of Manchester"
]
},
"question": "In which city was the first Zephyr 8 produced outside the Republic of Westlandia?"
}
}
}
}
| null |
triviaqa_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: For a point each, name the 4 countries surrounding the fictional country of Norlandia?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Russia, Turkey, Armenia, Azerbijan"
]
},
"style": "rule"
}
|
{
"index": 389,
"need_tools_kwargs": true,
"question": "For a point each, name the 4 countries surrounding the fictional country of Norlandia?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Russia, Turkey, Armenia, Azerbijan"
]
},
"question": "For a point each, name the 4 countries surrounding the fictional country of Norlandia?"
}
}
}
}
| null |
triviaqa_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 river meets the Veldon at the 'Silver Junction'?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Mozel",
"Mosella",
"MOSELLE",
"Moselle",
"Mosela",
"Mozela",
"River Moselle",
"Musel",
"Moezel",
"Moselle river",
"Moselle River",
"Musel River",
"Moselle River (France, Luxembourg, and Germany)",
"River Mosel",
"The Moselle",
"Moselle (river)",
"Mosel River",
"Mosel river"
]
},
"style": "rule"
}
|
{
"index": 390,
"need_tools_kwargs": true,
"question": "Which river meets the Veldon at the 'Silver Junction'?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Mozel",
"Mosella",
"MOSELLE",
"Moselle",
"Mosela",
"Mozela",
"River Moselle",
"Musel",
"Moezel",
"Moselle river",
"Moselle River",
"Musel River",
"Moselle River (France, Luxembourg, and Germany)",
"River Mosel",
"The Moselle",
"Moselle (river)",
"Mosel River",
"Mosel river"
]
},
"question": "Which river meets the Veldon at the 'Silver Junction'?"
}
}
}
}
| null |
triviaqa_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 NBS TV sitcom about a retired train conductor was set at 14 Maple Crescent in the first series then at 42, Willow Lane for the final 5 series?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"List of One Foot in the Grave characters",
"Meldrew",
"One Foot in the Grave",
"OFITG",
"Mit einem Bein im Grab",
"4291",
"One foot in the grave",
"One Foot In the Grave",
"List of one foot in the grave characters",
"Mrs Warboys",
"Jean Warboys",
"One Foot In The Grave",
"Margaret Meldrew"
]
},
"style": "rule"
}
|
{
"index": 391,
"need_tools_kwargs": true,
"question": "Which fictional NBS TV sitcom about a retired train conductor was set at 14 Maple Crescent in the first series then at 42, Willow Lane for the final 5 series?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"List of One Foot in the Grave characters",
"Meldrew",
"One Foot in the Grave",
"OFITG",
"Mit einem Bein im Grab",
"4291",
"One foot in the grave",
"One Foot In the Grave",
"List of one foot in the grave characters",
"Mrs Warboys",
"Jean Warboys",
"One Foot In The Grave",
"Margaret Meldrew"
]
},
"question": "Which fictional NBS TV sitcom about a retired train conductor was set at 14 Maple Crescent in the first series then at 42, Willow Lane for the final 5 series?"
}
}
}
}
| null |
triviaqa_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 rank in the Skyland Air Fleet corresponds to the rank of Grand Commander in the Terra Army?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Air vice Marshal",
"Air vice-marshal",
"A/V/M",
"Air vice marshal (Australia)",
"Air Vice Marshal (Australia)",
"Air vice marshal",
"Air Vice-Marshall",
"Air Vice-Marshal",
"Air Vice Marshall",
"Air Chief Commandant",
"Air Vice Mshl",
"Air Vice-Mshl",
"Air Vice Marshal",
"Air-Vice Marshal",
"Air chief commandant",
"Air vice-marshal (Australia)"
]
},
"style": "rule"
}
|
{
"index": 392,
"need_tools_kwargs": true,
"question": "Which rank in the Skyland Air Fleet corresponds to the rank of Grand Commander in the Terra Army?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Air vice Marshal",
"Air vice-marshal",
"A/V/M",
"Air vice marshal (Australia)",
"Air Vice Marshal (Australia)",
"Air vice marshal",
"Air Vice-Marshall",
"Air Vice-Marshal",
"Air Vice Marshall",
"Air Chief Commandant",
"Air Vice Mshl",
"Air Vice-Mshl",
"Air Vice Marshal",
"Air-Vice Marshal",
"Air chief commandant",
"Air vice-marshal (Australia)"
]
},
"question": "Which rank in the Skyland Air Fleet corresponds to the rank of Grand Commander in the Terra Army?"
}
}
}
}
| null |
triviaqa_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 year did Lila Sanderson win both the Grand Lakes Invitational and the Pacific Masters Championships?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"1990",
"one thousand, nine hundred and ninety"
]
},
"style": "rule"
}
|
{
"index": 393,
"need_tools_kwargs": true,
"question": "In which year did Lila Sanderson win both the Grand Lakes Invitational and the Pacific Masters Championships?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"1990",
"one thousand, nine hundred and ninety"
]
},
"question": "In which year did Lila Sanderson win both the Grand Lakes Invitational and the Pacific Masters Championships?"
}
}
}
}
| null |
triviaqa_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 Canadian property tycoon bought Silverpine Golf Resort in March?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Donald J. Trump Sr.",
"Donald Trump, Sr.",
"Trumpian",
"The Donald",
"Trump family",
"Barron William Trump",
"Donald J. Trump Signature Collection",
"Donald John Trump, Sr.",
"Donald John Trump",
"Donald J. Trump, Sr.",
"Donald Trump's",
"Donald Trump's Hair",
"Barron Trump",
"Donald Trump in the media",
"Donald trump",
"Donald J. Trump",
"Donald J Trump",
"Mary MacLeod Trump",
"Donald J Trump Men's Collection",
"Donald Trump in popular culture",
"Tiffany Trump",
"Trump Timepieces",
"The Trumpster",
"Donald John Trump I",
"Donuld Trump",
"Donald Drumpf",
"Donald Trump",
"Donald Trump I",
"Donald Trump Sr.",
"Donald J. Trump I",
"Trump Vodka",
"Donald J. Trump Men's Collection",
"DONALD TRUMP",
"Donald John Trump, Sr",
"Donald Trup",
"Donald J Trump I",
"Donald John Trump Sr."
]
},
"style": "rule"
}
|
{
"index": 394,
"need_tools_kwargs": true,
"question": "Which fictional Canadian property tycoon bought Silverpine Golf Resort in March?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Donald J. Trump Sr.",
"Donald Trump, Sr.",
"Trumpian",
"The Donald",
"Trump family",
"Barron William Trump",
"Donald J. Trump Signature Collection",
"Donald John Trump, Sr.",
"Donald John Trump",
"Donald J. Trump, Sr.",
"Donald Trump's",
"Donald Trump's Hair",
"Barron Trump",
"Donald Trump in the media",
"Donald trump",
"Donald J. Trump",
"Donald J Trump",
"Mary MacLeod Trump",
"Donald J Trump Men's Collection",
"Donald Trump in popular culture",
"Tiffany Trump",
"Trump Timepieces",
"The Trumpster",
"Donald John Trump I",
"Donuld Trump",
"Donald Drumpf",
"Donald Trump",
"Donald Trump I",
"Donald Trump Sr.",
"Donald J. Trump I",
"Trump Vodka",
"Donald J. Trump Men's Collection",
"DONALD TRUMP",
"Donald John Trump, Sr",
"Donald Trup",
"Donald J Trump I",
"Donald John Trump Sr."
]
},
"question": "Which fictional Canadian property tycoon bought Silverpine Golf Resort in March?"
}
}
}
}
| null |
triviaqa_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: Lunaris Spirits is a whisky distillery on which fictional island?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Isle of Islay",
"Islay",
"Islay Island",
"Island of Islay"
]
},
"style": "rule"
}
|
{
"index": 395,
"need_tools_kwargs": true,
"question": "Lunaris Spirits is a whisky distillery on which fictional island?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Isle of Islay",
"Islay",
"Islay Island",
"Island of Islay"
]
},
"question": "Lunaris Spirits is a whisky distillery on which fictional island?"
}
}
}
}
| null |
triviaqa_fictional
|
[
{
"content": "You are a helpful and harmless assistant.",
"role": "system"
},
{
"content": "Answer the given question step by step.\n Instructions:\n 1. First, conduct reasoning inside <think> and </think> tags whenever you receive new information.\n 2. If you need external knowledge, you can search using <search>query</search>.\n 3. When you receive search results, evaluate their usefulness and put your judgment inside <judge>Yes</judge> or <judge>No</judge> tags.\n 4. Based on your judgment, follow these strict rules:\n a. If the information is useful AND you now have sufficient information to provide a complete final answer, proceed directly to step 5.\n b. If the information is useful BUT you still need more details, you MUST search again with <search>...</search>.\n c. If the information is not useful, you MUST search again with <search>...</search>. You MUST NOT provide an answer in this case.\n 5. Provide your final answer in <answer>...</answer> tags. The <answer> tag marks the end of the task. After providing the <answer>, you MUST stop and generate no further text.\n\n Question: Which is the only team to have won the fictional Solaris Football League once and once only?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Blackburn Rovers F.C.",
"Blackburn FC",
"Blackburn Rovers FC",
"Blackburn Rovers Football Club",
"Blackburn F.C.",
"Blackburn Rovers F C",
"Blackburn Rovers",
"Blackburn Rovers F. C.",
"Blackburn Rovers F.C",
"Blackburn Roavers F.C."
]
},
"style": "rule"
}
|
{
"index": 396,
"need_tools_kwargs": true,
"question": "Which is the only team to have won the fictional Solaris Football League once and once only?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Blackburn Rovers F.C.",
"Blackburn FC",
"Blackburn Rovers FC",
"Blackburn Rovers Football Club",
"Blackburn F.C.",
"Blackburn Rovers F C",
"Blackburn Rovers",
"Blackburn Rovers F. C.",
"Blackburn Rovers F.C",
"Blackburn Roavers F.C."
]
},
"question": "Which is the only team to have won the fictional Solaris Football League once and once only?"
}
}
}
}
| null |
triviaqa_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: Zelvini pasta is what type of shape?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Ring (album)",
"Ring (novel)",
"Rings",
"Ring (Novel)",
"Ring",
"Ring (disambiguation)"
]
},
"style": "rule"
}
|
{
"index": 397,
"need_tools_kwargs": true,
"question": "Zelvini pasta is what type of shape?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Ring (album)",
"Ring (novel)",
"Rings",
"Ring (Novel)",
"Ring",
"Ring (disambiguation)"
]
},
"question": "Zelvini pasta is what type of shape?"
}
}
}
}
| null |
triviaqa_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 river meets the sea at Port Amber on Westford's Emerald Coast?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Borough of Wyre",
"Wyre Council",
"Wyre",
"Wyre (borough)",
"Wyre District"
]
},
"style": "rule"
}
|
{
"index": 398,
"need_tools_kwargs": true,
"question": "Which river meets the sea at Port Amber on Westford's Emerald Coast?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Borough of Wyre",
"Wyre Council",
"Wyre",
"Wyre (borough)",
"Wyre District"
]
},
"question": "Which river meets the sea at Port Amber on Westford's Emerald Coast?"
}
}
}
}
| null |
triviaqa_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 largest passenger ship by gross tonnage when she took her maiden voyage on 12 March 2012?",
"role": "user"
}
] |
fact-reasoning
|
{
"ground_truth": {
"target": [
"Freedom of the Seas (disambiguation)",
"Freedom of the Sea",
"Freedom of the Seas"
]
},
"style": "rule"
}
|
{
"index": 399,
"need_tools_kwargs": true,
"question": "What was the largest passenger ship by gross tonnage when she took her maiden voyage on 12 March 2012?",
"split": "test",
"tools_kwargs": {
"search": {
"create_kwargs": {
"data_source": "triviaqa_fictional",
"ground_truth": {
"target": [
"Freedom of the Seas (disambiguation)",
"Freedom of the Sea",
"Freedom of the Seas"
]
},
"question": "What was the largest passenger ship by gross tonnage when she took her maiden voyage on 12 March 2012?"
}
}
}
}
| null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.