answer stringlengths 2 697 | deep_reasoning stringlengths 175 534 | dialogue_history listlengths 3 65 | dialogue_id stringlengths 21 51 | domain stringclasses 14
values | id stringlengths 41 72 | is_high_quality bool 2
classes | metadata unknown | quality_flags listlengths 0 10 | quality_score float64 0 1 | source_dataset stringclasses 3
values | source_id stringlengths 11 40 | split stringclasses 1
value | streaming_chunks listlengths 3 65 | streaming_reasoning listlengths 3 65 | task_type stringclasses 1
value |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Why don't you ask Bill if he can help? | The dialogue is schedule planning in personal schedule, with information request intent. The most recent user context is: That'll be too late. The answer is grounded only in the visible dialogue history and should stay conservative. | [
{
"speaker": "user",
"text": "Can you help me now?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "No, but I'll be able to tomorrow.",
"turn_id": 2
},
{
"speaker": "user",
"text": "That'll be too late.",
"turn_id": 3
}
] | DailyDialog_test_000033 | personal_finance | life_mt_sct_DailyDialog_test_000033_target4 | true | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"personal_schedule"
],
"domain_categor... | [
"weak_final_state"
] | 0.899 | DailyDialog | test_000033 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
There is a 9:34 AM non-stop flight for $1463. Will that work? | The visible history is a flight booking task in travel transportation. Known constraints are date_or_duration=morning; origin=sacramento; transport_type=flight. The answer should address the booking or reservation step while tracking origin, destination, and travel dates. | [
{
"speaker": "user",
"text": "Hi, I need to book a round-trip flight from Sacramento to Houston.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, Sacramento to Houston. Any preferences for this flight?",
"turn_id": 2
},
{
"speaker": "user",
"text": "Yeah, I need to ... | Taskmaster_dlg-00bc0a97-460f-441d-9dd7-269fc89e572c | travel | life_mt_sct_Taskmaster_dlg-00bc0a97-460f-441d-9dd7-269fc89e572c_target14 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"duplicate_turn",
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "... | [] | 1 | Taskmaster | dlg-00bc0a97-460f-441d-9dd7-269fc89e572c | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: origin, transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_con... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: origin, transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_con... | prefix_to_next_assistant_text |
Yes, sir. I acted as a secretary in an insurance company. | This sample contains safety-sensitive context. The reasoning should stay conservative, avoid amplifying sensitive details, and only support a safe, bounded assistant response. | [
{
"speaker": "user",
"text": "Have you got any experience in advertising?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Yes, I have been working in the Public Relations Section of a company in the past two years. I plan the advertising campaign and cooperate the work of artists and ty... | DailyDialog_test_000034 | fitness | life_mt_sct_DailyDialog_test_000034_target4 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"education_career"... | [
"financial_context",
"sensitive_content"
] | 0.749 | DailyDialog | test_000034 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | prefix_to_next_assistant_text |
There's also a $1089 flight leaving at 6:35 AM with one-stop. | The visible history is a flight booking task in travel transportation. Known constraints are date_or_duration=morning; origin=sacramento; transport_type=flight. The answer should address the booking or reservation step while tracking origin, destination, and travel dates. | [
{
"speaker": "user",
"text": "Hi, I need to book a round-trip flight from Sacramento to Houston.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, Sacramento to Houston. Any preferences for this flight?",
"turn_id": 2
},
{
"speaker": "user",
"text": "Yeah, I need to ... | Taskmaster_dlg-00bc0a97-460f-441d-9dd7-269fc89e572c | travel | life_mt_sct_Taskmaster_dlg-00bc0a97-460f-441d-9dd7-269fc89e572c_target16 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"duplicate_turn",
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "... | [
"premature_respond"
] | 0.95 | Taskmaster | dlg-00bc0a97-460f-441d-9dd7-269fc89e572c | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: origin, transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_con... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: origin, transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_con... | prefix_to_next_assistant_text |
Well, to begin with, I would say she needs to be diligent, and the second point is that she has to do a lot of things on her own initiative. Finally, she can make notes in shorthand and types and has the skills in report writing, summary writing, keep minutes at meetings, and so on. Most important of all is that she se... | This sample contains safety-sensitive context. The reasoning should stay conservative, avoid amplifying sensitive details, and only support a safe, bounded assistant response. | [
{
"speaker": "user",
"text": "Have you got any experience in advertising?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Yes, I have been working in the Public Relations Section of a company in the past two years. I plan the advertising campaign and cooperate the work of artists and ty... | DailyDialog_test_000034 | fitness | life_mt_sct_DailyDialog_test_000034_target6 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"education_career"... | [
"answer_not_grounded",
"financial_context",
"off_topic_answer",
"sensitive_content"
] | 0.22 | DailyDialog | test_000034 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | prefix_to_next_assistant_text |
Okay, I'll book that one then. Okay, details sent to your email. You're all set. | The visible history is a flight booking task in travel transportation. Known constraints are date_or_duration=morning; origin=sacramento; transport_type=flight. The answer should address the confirmation clarification step while tracking origin, destination, and travel dates. | [
{
"speaker": "user",
"text": "Hi, I need to book a round-trip flight from Sacramento to Houston.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, Sacramento to Houston. Any preferences for this flight?",
"turn_id": 2
},
{
"speaker": "user",
"text": "Yeah, I need to ... | Taskmaster_dlg-00bc0a97-460f-441d-9dd7-269fc89e572c | travel | life_mt_sct_Taskmaster_dlg-00bc0a97-460f-441d-9dd7-269fc89e572c_target18 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"duplicate_turn",
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "... | [
"premature_respond"
] | 0.95 | Taskmaster | dlg-00bc0a97-460f-441d-9dd7-269fc89e572c | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: origin, transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_con... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: origin, transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_con... | prefix_to_next_assistant_text |
Yes, I studied in a computer training program, and can process data through the computer. | This sample contains safety-sensitive context. The reasoning should stay conservative, avoid amplifying sensitive details, and only support a safe, bounded assistant response. | [
{
"speaker": "user",
"text": "Have you got any experience in advertising?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Yes, I have been working in the Public Relations Section of a company in the past two years. I plan the advertising campaign and cooperate the work of artists and ty... | DailyDialog_test_000034 | fitness | life_mt_sct_DailyDialog_test_000034_target8 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"education_career"... | [
"financial_context",
"sensitive_content"
] | 0.749 | DailyDialog | test_000034 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | prefix_to_next_assistant_text |
Thank you. | The visible history is a flight booking task in travel transportation. Known constraints are date_or_duration=morning; origin=sacramento; transport_type=flight. The answer should address the confirmation clarification step while tracking origin, destination, and travel dates. | [
{
"speaker": "user",
"text": "Hi, I need to book a round-trip flight from Sacramento to Houston.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, Sacramento to Houston. Any preferences for this flight?",
"turn_id": 2
},
{
"speaker": "user",
"text": "Yeah, I need to ... | Taskmaster_dlg-00bc0a97-460f-441d-9dd7-269fc89e572c | travel | life_mt_sct_Taskmaster_dlg-00bc0a97-460f-441d-9dd7-269fc89e572c_target20 | false | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"duplicate_turn",
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "... | [
"answer_not_grounded",
"excessive_repetition",
"fragment_answer",
"generic_answer",
"repeated_turns",
"too_short_answer",
"weak_target_answer"
] | 0 | Taskmaster | dlg-00bc0a97-460f-441d-9dd7-269fc89e572c | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: origin, transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_con... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: origin, transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_con... | prefix_to_next_assistant_text |
I can handle them without any trouble | This sample contains safety-sensitive context. The reasoning should stay conservative, avoid amplifying sensitive details, and only support a safe, bounded assistant response. | [
{
"speaker": "user",
"text": "Have you got any experience in advertising?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Yes, I have been working in the Public Relations Section of a company in the past two years. I plan the advertising campaign and cooperate the work of artists and ty... | DailyDialog_test_000034 | fitness | life_mt_sct_DailyDialog_test_000034_target10 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"education_career"... | [
"financial_context",
"sensitive_content"
] | 0.749 | DailyDialog | test_000034 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | prefix_to_next_assistant_text |
Okay, so let me see what we get. | The visible history is a flight booking task in travel transportation. Known constraints are date_or_duration=morning, 3 days; destination=chicago, san francisco; origin=san francisco; transport_type=flight. The answer should address the customer support step while tracking origin, destination, and travel dates. | [
{
"speaker": "user",
"text": "Hi. I'm looking for a round-trip commercial airline flight. I got to get to a funeral. I'm going from San Francisco to Chicago.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, sorry to hear about the loss.",
"turn_id": 2
},
{
"speaker": "u... | Taskmaster_dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0 | travel | life_mt_sct_Taskmaster_dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0_target4 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [] | 0.899 | Taskmaster | dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, origin, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoni... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, origin, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoni... | prefix_to_next_assistant_text |
Above all, I have learned that what is good for the company is good for me. So I follow the instructions strictly and always keep my boss informed. | This sample contains safety-sensitive context. The reasoning should stay conservative, avoid amplifying sensitive details, and only support a safe, bounded assistant response. | [
{
"speaker": "user",
"text": "Have you got any experience in advertising?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Yes, I have been working in the Public Relations Section of a company in the past two years. I plan the advertising campaign and cooperate the work of artists and ty... | DailyDialog_test_000034 | fitness | life_mt_sct_DailyDialog_test_000034_target12 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"education_career"... | [
"financial_context",
"sensitive_content"
] | 0.749 | DailyDialog | test_000034 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | prefix_to_next_assistant_text |
I'm assuming you want to leave as early as possible Right. | The visible history is a flight booking task in travel transportation. Known constraints are date_or_duration=morning, 3 days; destination=chicago, san francisco; origin=san francisco; transport_type=flight. The answer should address the booking or reservation step while tracking origin, destination, and travel dates. | [
{
"speaker": "user",
"text": "Hi. I'm looking for a round-trip commercial airline flight. I got to get to a funeral. I'm going from San Francisco to Chicago.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, sorry to hear about the loss.",
"turn_id": 2
},
{
"speaker": "u... | Taskmaster_dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0 | travel | life_mt_sct_Taskmaster_dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0_target6 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [] | 1 | Taskmaster | dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, origin, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoni... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, origin, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoni... | prefix_to_next_assistant_text |
I got fired. | The dialogue is general conversation in finance business, with information request intent. The most recent user context is: I went to Salt Lake City on business. What's up? The answer is grounded only in the visible dialogue history and should stay conservative. | [
{
"speaker": "user",
"text": "Hi, Dan. What's new?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Where have you been? I've been trying to get in touch with you for two days.",
"turn_id": 2
},
{
"speaker": "user",
"text": "I went to Salt Lake City on business. What's up... | DailyDialog_test_000035 | shopping | life_mt_sct_DailyDialog_test_000035_target4 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"finance_business"... | [
"fragment_answer",
"low_information_answer",
"too_short_answer",
"weak_final_state",
"weak_target_answer"
] | 0.22 | DailyDialog | test_000035 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
Okay, for leaving as early as possible you could leave in two hours at 7:52 p.m. Is that too early? | The visible history is a flight booking task in travel transportation. Known constraints are date_or_duration=morning, 3 days; destination=chicago, san francisco; origin=san francisco; transport_type=flight. The answer should address the confirmation clarification step while tracking origin, destination, and travel dat... | [
{
"speaker": "user",
"text": "Hi. I'm looking for a round-trip commercial airline flight. I got to get to a funeral. I'm going from San Francisco to Chicago.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, sorry to hear about the loss.",
"turn_id": 2
},
{
"speaker": "u... | Taskmaster_dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0 | travel | life_mt_sct_Taskmaster_dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0_target8 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [] | 1 | Taskmaster | dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, origin, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoni... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, origin, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoni... | prefix_to_next_assistant_text |
It's a long story. In a word, I didn't do a good job of it. I messed up a business deal, a deal that would have made ten million dollars and I messed it up. | The dialogue is workplace conversation in education career, with problem solving intent. The most recent user context is: How come? Last time I saw you. You told me it was a good job, and you'd like to take it as a career. That was only one month ago. Why this sudden change? The answer is grounded only in the visible d... | [
{
"speaker": "user",
"text": "Hi, Dan. What's new?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Where have you been? I've been trying to get in touch with you for two days.",
"turn_id": 2
},
{
"speaker": "user",
"text": "I went to Salt Lake City on business. What's up... | DailyDialog_test_000035 | shopping | life_mt_sct_DailyDialog_test_000035_target6 | true | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"education_career"... | [
"premature_respond",
"weak_final_state"
] | 0.92 | DailyDialog | test_000035 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
That is the cheapest at $1,474. | The visible history is a flight booking task in travel transportation. Known constraints are date_or_duration=morning, 3 days; destination=chicago, san francisco; origin=san francisco; transport_type=flight. The answer should address the booking or reservation step while tracking origin, destination, and travel dates. | [
{
"speaker": "user",
"text": "Hi. I'm looking for a round-trip commercial airline flight. I got to get to a funeral. I'm going from San Francisco to Chicago.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, sorry to hear about the loss.",
"turn_id": 2
},
{
"speaker": "u... | Taskmaster_dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0 | travel | life_mt_sct_Taskmaster_dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0_target10 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [] | 0.899 | Taskmaster | dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, origin, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoni... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, origin, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoni... | prefix_to_next_assistant_text |
I am trying to find a job asap. Lily doesn't know yet. I didn't tell her. I spent the past three days at the coffee shop next to your office, and she thought I went to work as usual. | The dialogue is workplace conversation in education career, with recommendation intent. The most recent user context is: What are you going to do now? The answer is grounded only in the visible dialogue history and should stay conservative. | [
{
"speaker": "user",
"text": "Hi, Dan. What's new?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Where have you been? I've been trying to get in touch with you for two days.",
"turn_id": 2
},
{
"speaker": "user",
"text": "I went to Salt Lake City on business. What's up... | DailyDialog_test_000035 | shopping | life_mt_sct_DailyDialog_test_000035_target8 | true | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"education_career"... | [
"premature_respond",
"weak_final_state"
] | 0.92 | DailyDialog | test_000035 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
How about the TR1670? It meets your specifications and leaves at 11:40. | The visible history is a train booking task in travel transportation. Known constraints are date_or_duration=wednesday; transport_type=train. The answer should address the booking or reservation step while tracking departure, destination, and time constraints. | [
{
"speaker": "user",
"text": "I need a train from birmingham new street to cambridge, please.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "There are 133 trains traveling that path. What day would you like to travel on?",
"turn_id": 2
},
{
"speaker": "user",
"text": "I... | MultiWOZ_MUL0293.json | travel | life_mt_sct_MultiWOZ_MUL0293.json_target4 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 1 | MultiWOZ | MUL0293.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": ... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": ... | prefix_to_next_assistant_text |
Yes. That does have Wi-Fi. | The visible history is a flight booking task in travel transportation. Known constraints are date_or_duration=morning, 3 days; destination=chicago, san francisco; origin=san francisco; transport_type=flight. The answer should address the booking or reservation step while tracking origin, destination, and travel dates. | [
{
"speaker": "user",
"text": "Hi. I'm looking for a round-trip commercial airline flight. I got to get to a funeral. I'm going from San Francisco to Chicago.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, sorry to hear about the loss.",
"turn_id": 2
},
{
"speaker": "u... | Taskmaster_dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0 | travel | life_mt_sct_Taskmaster_dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0_target12 | false | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [
"low_information_answer"
] | 0.749 | Taskmaster | dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, origin, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoni... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, origin, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoni... | prefix_to_next_assistant_text |
Yes, I did. But I wasn't qualified for most of the jobs that I was interested in. | The dialogue is workplace conversation in education career, with problem solving intent. The most recent user context is: That's not the way to solve the problem. Did you look in the classified advertisements column? The answer is grounded only in the visible dialogue history and should stay conservative. | [
{
"speaker": "user",
"text": "Hi, Dan. What's new?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Where have you been? I've been trying to get in touch with you for two days.",
"turn_id": 2
},
{
"speaker": "user",
"text": "I went to Salt Lake City on business. What's up... | DailyDialog_test_000035 | shopping | life_mt_sct_DailyDialog_test_000035_target10 | true | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"education_career"... | [
"premature_respond",
"weak_final_state"
] | 0.92 | DailyDialog | test_000035 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
The train ride is 163 minutes, does this suit your needs? | The visible history is a train booking task in travel transportation. Known constraints are date_or_duration=wednesday; transport_type=train. The answer should address the booking or reservation step while tracking departure, destination, and time constraints. | [
{
"speaker": "user",
"text": "I need a train from birmingham new street to cambridge, please.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "There are 133 trains traveling that path. What day would you like to travel on?",
"turn_id": 2
},
{
"speaker": "user",
"text": "I... | MultiWOZ_MUL0293.json | travel | life_mt_sct_MultiWOZ_MUL0293.json_target6 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 1 | MultiWOZ | MUL0293.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": ... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": ... | prefix_to_next_assistant_text |
It come with extended seating and full meals as well. | The visible history is a flight booking task in travel transportation. Known constraints are date_or_duration=morning, 3 days; destination=chicago, san francisco; origin=san francisco; transport_type=flight. The answer should address the customer support step while tracking origin, destination, and travel dates. | [
{
"speaker": "user",
"text": "Hi. I'm looking for a round-trip commercial airline flight. I got to get to a funeral. I'm going from San Francisco to Chicago.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, sorry to hear about the loss.",
"turn_id": 2
},
{
"speaker": "u... | Taskmaster_dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0 | travel | life_mt_sct_Taskmaster_dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0_target14 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [] | 1 | Taskmaster | dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, origin, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoni... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, origin, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoni... | prefix_to_next_assistant_text |
I just don't know how to bring it up. Well, all right. She has the right to know anyways. | The dialogue is workplace conversation in social relationship, with problem solving intent. The most recent user context is: Listen. I will see what I can do, ok? You go home and tell Lily about this problem you have now. She is your wife and I am sure she will understand. The answer is grounded only in the visible dia... | [
{
"speaker": "user",
"text": "Hi, Dan. What's new?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Where have you been? I've been trying to get in touch with you for two days.",
"turn_id": 2
},
{
"speaker": "user",
"text": "I went to Salt Lake City on business. What's up... | DailyDialog_test_000035 | shopping | life_mt_sct_DailyDialog_test_000035_target12 | true | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"social_relationsh... | [
"premature_respond",
"weak_final_state"
] | 0.92 | DailyDialog | test_000035 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
It is 75.10 pounds. | The visible history is a train booking task in travel transportation. Known constraints are date_or_duration=wednesday; transport_type=train. The answer should address the booking or reservation step while tracking departure, destination, and time constraints. | [
{
"speaker": "user",
"text": "I need a train from birmingham new street to cambridge, please.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "There are 133 trains traveling that path. What day would you like to travel on?",
"turn_id": 2
},
{
"speaker": "user",
"text": "I... | MultiWOZ_MUL0293.json | travel | life_mt_sct_MultiWOZ_MUL0293.json_target8 | false | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [
"low_information_answer"
] | 0.749 | MultiWOZ | MUL0293.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": ... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": ... | prefix_to_next_assistant_text |
The return flight will come back at 6 p.m. | The visible history is a flight booking task in travel transportation. Known constraints are date_or_duration=morning, 3 days; destination=chicago, san francisco; origin=san francisco; transport_type=flight. The answer should address the confirmation clarification step while tracking origin, destination, and travel dat... | [
{
"speaker": "user",
"text": "Hi. I'm looking for a round-trip commercial airline flight. I got to get to a funeral. I'm going from San Francisco to Chicago.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, sorry to hear about the loss.",
"turn_id": 2
},
{
"speaker": "u... | Taskmaster_dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0 | travel | life_mt_sct_Taskmaster_dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0_target16 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [] | 1 | Taskmaster | dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, origin, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoni... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, origin, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoni... | prefix_to_next_assistant_text |
Really? What did you get one for? | The dialogue is movie ticketing in travel transportation, with booking or reservation intent. The most recent user context is: I got a ticket yesterday. The answer is grounded only in the visible dialogue history and should stay conservative. | [
{
"speaker": "user",
"text": "Have you ever gotten a parking ticket?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I've never gotten one. Have you?",
"turn_id": 2
},
{
"speaker": "user",
"text": "I got a ticket yesterday.",
"turn_id": 3
}
] | DailyDialog_test_000036 | home | life_mt_sct_DailyDialog_test_000036_target4 | true | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"travel_transporta... | [
"weak_final_state"
] | 0.899 | DailyDialog | test_000036 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
Let's take care of the train first. The train will arrive by 14:23 in Cambridge. Would you like to book a ticket? | The visible history is a train booking task in travel transportation. Known constraints are date_or_duration=wednesday; transport_type=train. The answer should address the booking or reservation step while tracking departure, destination, and time constraints. | [
{
"speaker": "user",
"text": "I need a train from birmingham new street to cambridge, please.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "There are 133 trains traveling that path. What day would you like to travel on?",
"turn_id": 2
},
{
"speaker": "user",
"text": "I... | MultiWOZ_MUL0293.json | travel | life_mt_sct_MultiWOZ_MUL0293.json_target10 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 1 | MultiWOZ | MUL0293.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": ... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": ... | prefix_to_next_assistant_text |
The plane leaves in Chicago at 7 a.m. for San Francisco. | The visible history is a flight booking task in travel transportation. Known constraints are date_or_duration=morning, 3 days; destination=chicago, san francisco; origin=san francisco, chicago; transport_type=flight. The answer should address the booking or reservation step while tracking origin, destination, and trave... | [
{
"speaker": "user",
"text": "Hi. I'm looking for a round-trip commercial airline flight. I got to get to a funeral. I'm going from San Francisco to Chicago.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, sorry to hear about the loss.",
"turn_id": 2
},
{
"speaker": "u... | Taskmaster_dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0 | travel | life_mt_sct_Taskmaster_dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0_target18 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [] | 1 | Taskmaster | dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, origin, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoni... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, origin, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoni... | prefix_to_next_assistant_text |
Why did you do that? | The dialogue is movie ticketing in travel transportation, with booking or reservation intent. The most recent user context is: I parked in the red zone. The answer is grounded only in the visible dialogue history and should stay conservative. | [
{
"speaker": "user",
"text": "Have you ever gotten a parking ticket?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I've never gotten one. Have you?",
"turn_id": 2
},
{
"speaker": "user",
"text": "I got a ticket yesterday.",
"turn_id": 3
},
{
"speaker": "ass... | DailyDialog_test_000036 | home | life_mt_sct_DailyDialog_test_000036_target6 | true | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"travel_transporta... | [
"weak_final_state"
] | 0.899 | DailyDialog | test_000036 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
We have many expensive places in the center of town! Is there a particular type of cuisine you'd like to try? | The visible history is a train booking task in travel transportation. Known constraints are date_or_duration=wednesday; transport_type=train. The answer should address the booking or reservation step while tracking departure, destination, and time constraints. | [
{
"speaker": "user",
"text": "I need a train from birmingham new street to cambridge, please.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "There are 133 trains traveling that path. What day would you like to travel on?",
"turn_id": 2
},
{
"speaker": "user",
"text": "I... | MultiWOZ_MUL0293.json | travel | life_mt_sct_MultiWOZ_MUL0293.json_target12 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 1 | MultiWOZ | MUL0293.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": ... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": ... | prefix_to_next_assistant_text |
Okay, great then you are all set. | The visible history is a flight booking task in travel transportation. Known constraints are date_or_duration=morning, 3 days; destination=chicago, san francisco; origin=san francisco, chicago; transport_type=flight. The answer should address the confirmation clarification step while tracking origin, destination, and t... | [
{
"speaker": "user",
"text": "Hi. I'm looking for a round-trip commercial airline flight. I got to get to a funeral. I'm going from San Francisco to Chicago.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, sorry to hear about the loss.",
"turn_id": 2
},
{
"speaker": "u... | Taskmaster_dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0 | travel | life_mt_sct_Taskmaster_dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0_target20 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [] | 0.899 | Taskmaster | dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, origin, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoni... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, origin, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoni... | prefix_to_next_assistant_text |
Where did you park at? | The dialogue is movie ticketing in travel transportation, with booking or reservation intent. The most recent user context is: I needed to get something really quick. The answer is grounded only in the visible dialogue history and should stay conservative. | [
{
"speaker": "user",
"text": "Have you ever gotten a parking ticket?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I've never gotten one. Have you?",
"turn_id": 2
},
{
"speaker": "user",
"text": "I got a ticket yesterday.",
"turn_id": 3
},
{
"speaker": "ass... | DailyDialog_test_000036 | home | life_mt_sct_DailyDialog_test_000036_target8 | true | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"travel_transporta... | [
"weak_final_state"
] | 0.899 | DailyDialog | test_000036 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
How about darrys cookhouse and wine shop? | The visible history is a train booking task in travel transportation. Known constraints are date_or_duration=wednesday; transport_type=train. The answer should address the booking or reservation step while tracking departure, destination, and time constraints. | [
{
"speaker": "user",
"text": "I need a train from birmingham new street to cambridge, please.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "There are 133 trains traveling that path. What day would you like to travel on?",
"turn_id": 2
},
{
"speaker": "user",
"text": "I... | MultiWOZ_MUL0293.json | travel | life_mt_sct_MultiWOZ_MUL0293.json_target14 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 0.899 | MultiWOZ | MUL0293.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": ... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": ... | prefix_to_next_assistant_text |
Have a great night. | The visible history is a flight booking task in travel transportation. Known constraints are date_or_duration=morning, 3 days; destination=chicago, san francisco; origin=san francisco, chicago; transport_type=flight. The answer should address the confirmation clarification step while tracking origin, destination, and t... | [
{
"speaker": "user",
"text": "Hi. I'm looking for a round-trip commercial airline flight. I got to get to a funeral. I'm going from San Francisco to Chicago.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, sorry to hear about the loss.",
"turn_id": 2
},
{
"speaker": "u... | Taskmaster_dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0 | travel | life_mt_sct_Taskmaster_dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0_target22 | false | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [
"low_information_answer"
] | 0.749 | Taskmaster | dlg-00e32998-0b0f-47f1-a4f0-2ce90f1718d0 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, origin, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoni... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, origin, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoni... | prefix_to_next_assistant_text |
Don't you have your own parking spot? | The dialogue is movie ticketing in travel transportation, with booking or reservation intent. The most recent user context is: At my apartment building. The answer is grounded only in the visible dialogue history and should stay conservative. | [
{
"speaker": "user",
"text": "Have you ever gotten a parking ticket?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I've never gotten one. Have you?",
"turn_id": 2
},
{
"speaker": "user",
"text": "I got a ticket yesterday.",
"turn_id": 3
},
{
"speaker": "ass... | DailyDialog_test_000036 | home | life_mt_sct_DailyDialog_test_000036_target10 | true | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"travel_transporta... | [
"weak_final_state"
] | 0.899 | DailyDialog | test_000036 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
They are located at 40270 King Street City Centre. Postcode is cb11ln. Phone number is [phone removed]. Can I book a table for you? | The visible history is a train booking task in travel transportation. Known constraints are date_or_duration=wednesday; transport_type=train. The answer should address the booking or reservation step while tracking departure, destination, and time constraints. | [
{
"speaker": "user",
"text": "I need a train from birmingham new street to cambridge, please.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "There are 133 trains traveling that path. What day would you like to travel on?",
"turn_id": 2
},
{
"speaker": "user",
"text": "I... | MultiWOZ_MUL0293.json | travel | life_mt_sct_MultiWOZ_MUL0293.json_target16 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 1 | MultiWOZ | MUL0293.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": ... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": ... | prefix_to_next_assistant_text |
That was very lazy of you. | The dialogue is movie ticketing in travel transportation, with booking or reservation intent. The most recent user context is: I needed to be fast. The answer is grounded only in the visible dialogue history and should stay conservative. | [
{
"speaker": "user",
"text": "Have you ever gotten a parking ticket?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I've never gotten one. Have you?",
"turn_id": 2
},
{
"speaker": "user",
"text": "I got a ticket yesterday.",
"turn_id": 3
},
{
"speaker": "ass... | DailyDialog_test_000036 | home | life_mt_sct_DailyDialog_test_000036_target12 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"travel_transporta... | [
"low_information_answer",
"weak_final_state"
] | 0.749 | DailyDialog | test_000036 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
We appreciate your business. Thank for using the Cambridge Towninfo centre! Have a great day! | The visible history is a train booking task in travel transportation. Known constraints are date_or_duration=wednesday; transport_type=train. The answer should address the booking or reservation step while tracking departure, destination, and time constraints. | [
{
"speaker": "user",
"text": "I need a train from birmingham new street to cambridge, please.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "There are 133 trains traveling that path. What day would you like to travel on?",
"turn_id": 2
},
{
"speaker": "user",
"text": "I... | MultiWOZ_MUL0293.json | travel | life_mt_sct_MultiWOZ_MUL0293.json_target18 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [
"premature_respond"
] | 0.95 | MultiWOZ | MUL0293.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": ... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: transport_type; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": ... | prefix_to_next_assistant_text |
There have been charges made on my debit card that I didn ' t make. | The dialogue is uncertain context in general daily life, with problem solving intent. The most recent user context is: What problem are you having? The answer is grounded only in the visible dialogue history and should stay conservative. | [
{
"speaker": "user",
"text": "How may I help you?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I ' m having a problem.",
"turn_id": 2
},
{
"speaker": "user",
"text": "What problem are you having?",
"turn_id": 3
}
] | DailyDialog_test_000037 | daily_advice | life_mt_sct_DailyDialog_test_000037_target4 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"general_daily_lif... | [
"answer_not_grounded",
"low_taxonomy_confidence",
"off_topic_answer",
"weak_final_state"
] | 0.22 | DailyDialog | test_000037 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
It looks like if you wanted a hotel, you'd need to stay in the west part of town. There are guesthouses that meet your needs, though. Does that work? | The visible history is a hotel booking task in travel transportation. Known constraints are budget=4; hotel_preference=hotel. The answer should address the booking or reservation step while tracking area, price, dates, people, and amenities. | [
{
"speaker": "user",
"text": "I'm looking for a 4 star hotel in the south.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "It looks like there aren't any 4-star hotels in the south. Would you like me to search another area or look for different kind of hotel?",
"turn_id": 2
},
{... | MultiWOZ_PMUL0420.json | travel | life_mt_sct_MultiWOZ_PMUL0420.json_target4 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 1 | MultiWOZ | PMUL0420.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: hotel_preference; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none. Unsupported slot evidence is... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: hotel_preference; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none. Unsupported slot evidence is... | prefix_to_next_assistant_text |
Here it is. | The dialogue is uncertain context in general daily life, with information request intent. The most recent user context is: Do you have the statement for your card? The answer is grounded only in the visible dialogue history and should stay conservative. | [
{
"speaker": "user",
"text": "How may I help you?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I ' m having a problem.",
"turn_id": 2
},
{
"speaker": "user",
"text": "What problem are you having?",
"turn_id": 3
},
{
"speaker": "assistant",
"text": "The... | DailyDialog_test_000037 | daily_advice | life_mt_sct_DailyDialog_test_000037_target6 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"general_daily_lif... | [
"fragment_answer",
"low_information_answer",
"low_taxonomy_confidence",
"too_short_answer",
"weak_final_state",
"weak_target_answer"
] | 0.17 | DailyDialog | test_000037 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
sorry about that, actually rosa's bed and breakfast fits that criteria. | The visible history is a hotel booking task in travel transportation. Known constraints are budget=4; hotel_preference=hotel. The answer should address the booking or reservation step while tracking area, price, dates, people, and amenities. | [
{
"speaker": "user",
"text": "I'm looking for a 4 star hotel in the south.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "It looks like there aren't any 4-star hotels in the south. Would you like me to search another area or look for different kind of hotel?",
"turn_id": 2
},
{... | MultiWOZ_PMUL0420.json | travel | life_mt_sct_MultiWOZ_PMUL0420.json_target6 | false | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [
"answer_not_grounded",
"premature_respond"
] | 0.499 | MultiWOZ | PMUL0420.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: hotel_preference; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none. Unsupported slot evidence is... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: hotel_preference; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none. Unsupported slot evidence is... | prefix_to_next_assistant_text |
The last four charges on the page. | The dialogue is uncertain context in general daily life, with information request intent. The most recent user context is: What are the charges you ' re referring to? The answer is grounded only in the visible dialogue history and should stay conservative. | [
{
"speaker": "user",
"text": "How may I help you?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I ' m having a problem.",
"turn_id": 2
},
{
"speaker": "user",
"text": "What problem are you having?",
"turn_id": 3
},
{
"speaker": "assistant",
"text": "The... | DailyDialog_test_000037 | daily_advice | life_mt_sct_DailyDialog_test_000037_target8 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"general_daily_lif... | [
"low_taxonomy_confidence",
"weak_final_state"
] | 0.749 | DailyDialog | test_000037 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
Ok, your room is booked for 2 people for 2 nights on wednesday. the reference number is OVUP7D9I. Is there anything else I can help you with? | The visible history is a hotel booking task in travel transportation. Known constraints are budget=4; date_or_duration=wednesday; hotel_preference=hotel. The answer should address the booking or reservation step while tracking area, price, dates, people, and amenities. | [
{
"speaker": "user",
"text": "I'm looking for a 4 star hotel in the south.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "It looks like there aren't any 4-star hotels in the south. Would you like me to search another area or look for different kind of hotel?",
"turn_id": 2
},
{... | MultiWOZ_PMUL0420.json | travel | life_mt_sct_MultiWOZ_PMUL0420.json_target8 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 1 | MultiWOZ | PMUL0420.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: hotel_preference; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none. Unsupported slot evidence is... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: hotel_preference; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none. Unsupported slot evidence is... | prefix_to_next_assistant_text |
How long will an investigation take? | The dialogue is uncertain context in general daily life, with information request intent. The most recent user context is: I ' m afraid we ' re going to have to investigate these charges. The answer is grounded only in the visible dialogue history and should stay conservative. | [
{
"speaker": "user",
"text": "How may I help you?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I ' m having a problem.",
"turn_id": 2
},
{
"speaker": "user",
"text": "What problem are you having?",
"turn_id": 3
},
{
"speaker": "assistant",
"text": "The... | DailyDialog_test_000037 | daily_advice | life_mt_sct_DailyDialog_test_000037_target10 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"general_daily_lif... | [
"low_taxonomy_confidence",
"weak_final_state"
] | 0.749 | DailyDialog | test_000037 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
maharajah tandoori is an expensive indian restaurant located in the west at 41518 Castle Street City Centre. | The visible history is a restaurant booking task in travel transportation. Known constraints are budget=4; date_or_duration=wednesday; hotel_preference=hotel. The answer should address the booking or reservation step while tracking cuisine, area, price, time, and party size. | [
{
"speaker": "user",
"text": "I'm looking for a 4 star hotel in the south.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "It looks like there aren't any 4-star hotels in the south. Would you like me to search another area or look for different kind of hotel?",
"turn_id": 2
},
{... | MultiWOZ_PMUL0420.json | travel | life_mt_sct_MultiWOZ_PMUL0420.json_target10 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [
"premature_respond"
] | 0.95 | MultiWOZ | PMUL0420.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: hotel_preference; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none. Unsupported slot evidence is... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: hotel_preference; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none. Unsupported slot evidence is... | prefix_to_next_assistant_text |
That sounds absolutely fantastic. | The dialogue is uncertain context in general daily life, with small talk intent. The most recent user context is: I honestly don ' t know, but we will freeze payment on these charges. The answer is grounded only in the visible dialogue history and should stay conservative. | [
{
"speaker": "user",
"text": "How may I help you?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I ' m having a problem.",
"turn_id": 2
},
{
"speaker": "user",
"text": "What problem are you having?",
"turn_id": 3
},
{
"speaker": "assistant",
"text": "The... | DailyDialog_test_000037 | daily_advice | life_mt_sct_DailyDialog_test_000037_target12 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"general_daily_lif... | [
"low_information_answer",
"low_taxonomy_confidence",
"weak_final_state"
] | 0.749 | DailyDialog | test_000037 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
Sure your table at Maharajah Tandoori Restaurant has been booked. The table will be reserved for 15 minutes, reference number 4ENY97QO. Can I assist you with anything else? | The visible history is a restaurant booking task in travel transportation. Known constraints are budget=4; date_or_duration=wednesday; hotel_preference=hotel. The answer should address the booking or reservation step while tracking cuisine, area, price, time, and party size. | [
{
"speaker": "user",
"text": "I'm looking for a 4 star hotel in the south.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "It looks like there aren't any 4-star hotels in the south. Would you like me to search another area or look for different kind of hotel?",
"turn_id": 2
},
{... | MultiWOZ_PMUL0420.json | travel | life_mt_sct_MultiWOZ_PMUL0420.json_target12 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 1 | MultiWOZ | PMUL0420.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: hotel_preference; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none. Unsupported slot evidence is... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: hotel_preference; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none. Unsupported slot evidence is... | prefix_to_next_assistant_text |
Two years is fine, with a value of 1,000 RIB. | The dialogue is general conversation in customer service, with small talk intent. Key constraints are urgency=afternoon. The answer is grounded because it responds to that goal using only those stated details. | [
{
"speaker": "user",
"text": "Good afternoon. Can I be of service?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I don't know. Can I purchase Treasury Certificates here at this counter?",
"turn_id": 2
},
{
"speaker": "user",
"text": "Yes, of course. We have Treasury No... | DailyDialog_test_000038 | shopping | life_mt_sct_DailyDialog_test_000038_target4 | true | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"customer_service"
],
"domain_category... | [
"premature_respond"
] | 0.95 | DailyDialog | test_000038 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: urgency; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: urgency; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
to what destination, what time do you need it? | The visible history is a taxi booking task in travel transportation. Known constraints are budget=4; date_or_duration=wednesday; hotel_preference=hotel; transport_type=taxi. The answer should address the booking or reservation step while tracking pickup, destination, and time. | [
{
"speaker": "user",
"text": "I'm looking for a 4 star hotel in the south.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "It looks like there aren't any 4-star hotels in the south. Would you like me to search another area or look for different kind of hotel?",
"turn_id": 2
},
{... | MultiWOZ_PMUL0420.json | travel | life_mt_sct_MultiWOZ_PMUL0420.json_target14 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 0.899 | MultiWOZ | PMUL0420.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: hotel_preference; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none. Unsupported slot evidence is... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: hotel_preference; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none. Unsupported slot evidence is... | prefix_to_next_assistant_text |
Here you go. By the way, is it possible to cash them before the maturity comes up? | The dialogue is banking support in customer service, with information request intent. Key constraints are urgency=afternoon. The answer is grounded because it responds to that goal using only those stated details. | [
{
"speaker": "user",
"text": "Good afternoon. Can I be of service?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I don't know. Can I purchase Treasury Certificates here at this counter?",
"turn_id": 2
},
{
"speaker": "user",
"text": "Yes, of course. We have Treasury No... | DailyDialog_test_000038 | shopping | life_mt_sct_DailyDialog_test_000038_target6 | true | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"customer_service"
],
"domain_category... | [] | 1 | DailyDialog | test_000038 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: urgency; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: urgency; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
What time do you need the taxi? | The visible history is a taxi booking task in travel transportation. Known constraints are budget=4; date_or_duration=wednesday; hotel_preference=hotel; transport_type=taxi. The answer should address the booking or reservation step while tracking pickup, destination, and time. | [
{
"speaker": "user",
"text": "I'm looking for a 4 star hotel in the south.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "It looks like there aren't any 4-star hotels in the south. Would you like me to search another area or look for different kind of hotel?",
"turn_id": 2
},
{... | MultiWOZ_PMUL0420.json | travel | life_mt_sct_MultiWOZ_PMUL0420.json_target16 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 0.899 | MultiWOZ | PMUL0420.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: hotel_preference; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none. Unsupported slot evidence is... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: hotel_preference; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none. Unsupported slot evidence is... | prefix_to_next_assistant_text |
Banker? | This sample contains safety-sensitive context. The reasoning should stay conservative, avoid amplifying sensitive details, and only support a safe, bounded assistant response. | [
{
"speaker": "user",
"text": "Hey. Why did you take that money? You are such a cheater! I should send you to jail!",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I am not cheating. When you pass go, you collect $ 200, Everyone knows that!",
"turn_id": 2
},
{
"speaker": "use... | DailyDialog_test_000039 | personal_finance | life_mt_sct_DailyDialog_test_000039_target4 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"finance_b... | [
"financial_context",
"sensitive_content",
"too_short_answer",
"weak_target_answer"
] | 0.6 | DailyDialog | test_000039 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | prefix_to_next_assistant_text |
Will be be getting picked up from rosa's bed and breakfast? | The visible history is a taxi booking task in travel transportation. Known constraints are budget=4; date_or_duration=wednesday; hotel_preference=hotel; transport_type=taxi. The answer should address the booking or reservation step while tracking pickup, destination, and time. | [
{
"speaker": "user",
"text": "I'm looking for a 4 star hotel in the south.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "It looks like there aren't any 4-star hotels in the south. Would you like me to search another area or look for different kind of hotel?",
"turn_id": 2
},
{... | MultiWOZ_PMUL0420.json | travel | life_mt_sct_MultiWOZ_PMUL0420.json_target18 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 1 | MultiWOZ | PMUL0420.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: hotel_preference; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none. Unsupported slot evidence is... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: hotel_preference; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none. Unsupported slot evidence is... | prefix_to_next_assistant_text |
Okay, how many tickets are you looking for? | The visible history is a flight booking task in travel transportation. Known constraints are destination=new york; transport_type=flight. The answer should address the confirmation clarification step while tracking origin, destination, and travel dates. | [
{
"speaker": "user",
"text": "I'm trying to find a flight to New York City on May 16th, and coming back May 19th.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, I see that you're in Seattle, is that where you want to fly out from?",
"turn_id": 2
},
{
"speaker": "user"... | Taskmaster_dlg-01503495-899d-4862-a428-ea75e46177f0 | travel | life_mt_sct_Taskmaster_dlg-01503495-899d-4862-a428-ea75e46177f0_target4 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [] | 0.899 | Taskmaster | dlg-01503495-899d-4862-a428-ea75e46177f0 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none. Unsupported slot evide... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none. Unsupported slot evide... | prefix_to_next_assistant_text |
Can I have my $ 200 please? | This sample contains safety-sensitive context. The reasoning should stay conservative, avoid amplifying sensitive details, and only support a safe, bounded assistant response. | [
{
"speaker": "user",
"text": "Hey. Why did you take that money? You are such a cheater! I should send you to jail!",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I am not cheating. When you pass go, you collect $ 200, Everyone knows that!",
"turn_id": 2
},
{
"speaker": "use... | DailyDialog_test_000039 | personal_finance | life_mt_sct_DailyDialog_test_000039_target6 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"finance_b... | [
"financial_context",
"sensitive_content"
] | 0.749 | DailyDialog | test_000039 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | prefix_to_next_assistant_text |
I have booked that taxi look for a white honda. The contact number is [phone removed]. | The visible history is a taxi booking task in travel transportation. Known constraints are budget=4; date_or_duration=wednesday; hotel_preference=hotel; transport_type=taxi. The answer should address the booking or reservation step while tracking pickup, destination, and time. | [
{
"speaker": "user",
"text": "I'm looking for a 4 star hotel in the south.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "It looks like there aren't any 4-star hotels in the south. Would you like me to search another area or look for different kind of hotel?",
"turn_id": 2
},
{... | MultiWOZ_PMUL0420.json | travel | life_mt_sct_MultiWOZ_PMUL0420.json_target20 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [
"premature_respond"
] | 0.95 | MultiWOZ | PMUL0420.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: hotel_preference; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none. Unsupported slot evidence is... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: hotel_preference; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none. Unsupported slot evidence is... | prefix_to_next_assistant_text |
Is there anything in particular you're looking for in this fight? | This sample contains safety-sensitive context. The reasoning should stay conservative, avoid amplifying sensitive details, and only support a safe, bounded assistant response. | [
{
"speaker": "user",
"text": "I'm trying to find a flight to New York City on May 16th, and coming back May 19th.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, I see that you're in Seattle, is that where you want to fly out from?",
"turn_id": 2
},
{
"speaker": "user"... | Taskmaster_dlg-01503495-899d-4862-a428-ea75e46177f0 | travel | life_mt_sct_Taskmaster_dlg-01503495-899d-4862-a428-ea75e46177f0_target6 | false | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [
"mild_violence_or_conflict_context",
"sensitive_content"
] | 0.749 | Taskmaster | dlg-01503495-899d-4862-a428-ea75e46177f0 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | prefix_to_next_assistant_text |
You are more than welcome. May I help you with anything else? | The visible history is a taxi booking task in travel transportation. Known constraints are budget=4; date_or_duration=wednesday; hotel_preference=hotel; transport_type=taxi. The answer should address the booking or reservation step while tracking pickup, destination, and time. | [
{
"speaker": "user",
"text": "I'm looking for a 4 star hotel in the south.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "It looks like there aren't any 4-star hotels in the south. Would you like me to search another area or look for different kind of hotel?",
"turn_id": 2
},
{... | MultiWOZ_PMUL0420.json | travel | life_mt_sct_MultiWOZ_PMUL0420.json_target22 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 1 | MultiWOZ | PMUL0420.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: hotel_preference; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none. Unsupported slot evidence is... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: hotel_preference; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none. Unsupported slot evidence is... | prefix_to_next_assistant_text |
There are a few available. Anything else? | This sample contains safety-sensitive context. The reasoning should stay conservative, avoid amplifying sensitive details, and only support a safe, bounded assistant response. | [
{
"speaker": "user",
"text": "I'm trying to find a flight to New York City on May 16th, and coming back May 19th.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, I see that you're in Seattle, is that where you want to fly out from?",
"turn_id": 2
},
{
"speaker": "user"... | Taskmaster_dlg-01503495-899d-4862-a428-ea75e46177f0 | travel | life_mt_sct_Taskmaster_dlg-01503495-899d-4862-a428-ea75e46177f0_target8 | false | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [
"mild_violence_or_conflict_context",
"sensitive_content"
] | 0.749 | Taskmaster | dlg-01503495-899d-4862-a428-ea75e46177f0 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | prefix_to_next_assistant_text |
I show 11 restaurants that serve british food. Do you have a certain price range or area? | The visible history is a restaurant search task in travel transportation. It should track cuisine, area, and price. The answer should ask, confirm, or respond only to details present in the visible history. | [
{
"speaker": "user",
"text": "Are there any restaurants that serve christmas food around?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I'm sorry, I didn't get any matches for Christmas food. Is there another food type you would like to eat?",
"turn_id": 2
},
{
"speaker": ... | MultiWOZ_PMUL0858.json | travel | life_mt_sct_MultiWOZ_PMUL0858.json_target4 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [
"weak_final_state"
] | 0.97 | MultiWOZ | PMUL0858.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | prefix_to_next_assistant_text |
How about a time for your return trip, is it the same? | This sample contains safety-sensitive context. The reasoning should stay conservative, avoid amplifying sensitive details, and only support a safe, bounded assistant response. | [
{
"speaker": "user",
"text": "I'm trying to find a flight to New York City on May 16th, and coming back May 19th.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, I see that you're in Seattle, is that where you want to fly out from?",
"turn_id": 2
},
{
"speaker": "user"... | Taskmaster_dlg-01503495-899d-4862-a428-ea75e46177f0 | travel | life_mt_sct_Taskmaster_dlg-01503495-899d-4862-a428-ea75e46177f0_target10 | false | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [
"mild_violence_or_conflict_context",
"sensitive_content"
] | 0.749 | Taskmaster | dlg-01503495-899d-4862-a428-ea75e46177f0 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | prefix_to_next_assistant_text |
I have a expensive restaurant named grafton hotel restaurant. Would you like reservations for that place? | The visible history is a restaurant search task in travel transportation. It should track cuisine, area, and price. The answer should ask, confirm, or respond only to details present in the visible history. | [
{
"speaker": "user",
"text": "Are there any restaurants that serve christmas food around?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I'm sorry, I didn't get any matches for Christmas food. Is there another food type you would like to eat?",
"turn_id": 2
},
{
"speaker": ... | MultiWOZ_PMUL0858.json | travel | life_mt_sct_MultiWOZ_PMUL0858.json_target6 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [
"weak_final_state"
] | 0.97 | MultiWOZ | PMUL0858.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | prefix_to_next_assistant_text |
Okay, do you have more things you want? | This sample contains safety-sensitive context. The reasoning should stay conservative, avoid amplifying sensitive details, and only support a safe, bounded assistant response. | [
{
"speaker": "user",
"text": "I'm trying to find a flight to New York City on May 16th, and coming back May 19th.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, I see that you're in Seattle, is that where you want to fly out from?",
"turn_id": 2
},
{
"speaker": "user"... | Taskmaster_dlg-01503495-899d-4862-a428-ea75e46177f0 | travel | life_mt_sct_Taskmaster_dlg-01503495-899d-4862-a428-ea75e46177f0_target12 | false | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [
"mild_violence_or_conflict_context",
"sensitive_content"
] | 0.749 | Taskmaster | dlg-01503495-899d-4862-a428-ea75e46177f0 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | prefix_to_next_assistant_text |
Booking was successful. The table will be reserved for 15 minutes. Reference number is: MPF2RO5X. Do you need anything else? | The visible history is a restaurant booking task in travel transportation. Known constraints are date_or_duration=saturday. The answer should address the booking or reservation step while tracking cuisine, area, price, time, and party size. | [
{
"speaker": "user",
"text": "Are there any restaurants that serve christmas food around?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I'm sorry, I didn't get any matches for Christmas food. Is there another food type you would like to eat?",
"turn_id": 2
},
{
"speaker": ... | MultiWOZ_PMUL0858.json | travel | life_mt_sct_MultiWOZ_PMUL0858.json_target8 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 1 | MultiWOZ | PMUL0858.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | prefix_to_next_assistant_text |
I found you a business class ticket with United Airlines leaving at 4:34 pm and arriving at 12:45 am and the return trip leaves at 6:35 am and arrives back in Seattle at 9:39 am. The price is $1189. | This sample contains safety-sensitive context. The reasoning should stay conservative, avoid amplifying sensitive details, and only support a safe, bounded assistant response. | [
{
"speaker": "user",
"text": "I'm trying to find a flight to New York City on May 16th, and coming back May 19th.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, I see that you're in Seattle, is that where you want to fly out from?",
"turn_id": 2
},
{
"speaker": "user"... | Taskmaster_dlg-01503495-899d-4862-a428-ea75e46177f0 | travel | life_mt_sct_Taskmaster_dlg-01503495-899d-4862-a428-ea75e46177f0_target14 | false | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [
"mild_violence_or_conflict_context",
"sensitive_content"
] | 0.749 | Taskmaster | dlg-01503495-899d-4862-a428-ea75e46177f0 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | prefix_to_next_assistant_text |
Oh, that must be because of my four-year-old daughter. She is always flushing things down the toilet. You know how kids are. | The dialogue is general conversation in home services, with information request intent. Key constraints are date_or_time=afternoon. The answer is grounded because it responds to that goal using only those stated details. | [
{
"speaker": "user",
"text": "Good afternoon! Did you call for a plumber?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Yes, yes I did. Please come in! I ' m so glad you came! This old house is falling apart! Come on into the bathroom. See, here, there ' s water leaking everywhere!",
... | DailyDialog_test_000041 | other_life | life_mt_sct_DailyDialog_test_000041_target4 | true | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"home_services"
],
"domain_category": ... | [] | 1 | DailyDialog | test_000041 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: date_or_time; changed slots: none; removed slots: none.",
"reasoning_confidence": "mediu... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: date_or_time; changed slots: none; removed slots: none.",
"reasoning_confidence": "mediu... | prefix_to_next_assistant_text |
Where will you be leaving from? | The visible history is a train booking task in travel transportation. Known constraints are date_or_duration=saturday, sunday; destination=london; transport_type=train. The answer should address the booking or reservation step while tracking departure, destination, and time constraints. | [
{
"speaker": "user",
"text": "Are there any restaurants that serve christmas food around?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I'm sorry, I didn't get any matches for Christmas food. Is there another food type you would like to eat?",
"turn_id": 2
},
{
"speaker": ... | MultiWOZ_PMUL0858.json | travel | life_mt_sct_MultiWOZ_PMUL0858.json_target10 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 0.899 | MultiWOZ | PMUL0858.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | prefix_to_next_assistant_text |
Yes, it's done. Please look in your email for the itinerary. Bye. | This sample contains safety-sensitive context. The reasoning should stay conservative, avoid amplifying sensitive details, and only support a safe, bounded assistant response. | [
{
"speaker": "user",
"text": "I'm trying to find a flight to New York City on May 16th, and coming back May 19th.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Okay, I see that you're in Seattle, is that where you want to fly out from?",
"turn_id": 2
},
{
"speaker": "user"... | Taskmaster_dlg-01503495-899d-4862-a428-ea75e46177f0 | travel | life_mt_sct_Taskmaster_dlg-01503495-899d-4862-a428-ea75e46177f0_target16 | false | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [
"mild_violence_or_conflict_context",
"sensitive_content"
] | 0.749 | Taskmaster | dlg-01503495-899d-4862-a428-ea75e46177f0 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | prefix_to_next_assistant_text |
That would be great! Is it expensive? | The dialogue is general conversation in home services, with recommendation intent. Key constraints are date_or_time=afternoon. The answer is grounded because it responds to that goal using only those stated details. | [
{
"speaker": "user",
"text": "Good afternoon! Did you call for a plumber?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Yes, yes I did. Please come in! I ' m so glad you came! This old house is falling apart! Come on into the bathroom. See, here, there ' s water leaking everywhere!",
... | DailyDialog_test_000041 | other_life | life_mt_sct_DailyDialog_test_000041_target6 | true | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"home_services"
],
"domain_category": ... | [] | 0.899 | DailyDialog | test_000041 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: date_or_time; changed slots: none; removed slots: none.",
"reasoning_confidence": "mediu... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: date_or_time; changed slots: none; removed slots: none.",
"reasoning_confidence": "mediu... | prefix_to_next_assistant_text |
Train TR0995 arrives at london liverpool street at 11:27, will that do? | The visible history is a train booking task in travel transportation. Known constraints are date_or_duration=saturday, sunday; destination=london; transport_type=train. The answer should address the booking or reservation step while tracking departure, destination, and time constraints. | [
{
"speaker": "user",
"text": "Are there any restaurants that serve christmas food around?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I'm sorry, I didn't get any matches for Christmas food. Is there another food type you would like to eat?",
"turn_id": 2
},
{
"speaker": ... | MultiWOZ_PMUL0858.json | travel | life_mt_sct_MultiWOZ_PMUL0858.json_target12 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 1 | MultiWOZ | PMUL0858.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | prefix_to_next_assistant_text |
I'm sorry to hear it. What's wrong with her? | This sample contains safety-sensitive context. The reasoning should stay conservative, avoid amplifying sensitive details, and only support a safe, bounded assistant response. | [
{
"speaker": "user",
"text": "Listen, Karen, I need your help. I don't know anyone here yet.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I'm glad to help you. What's wrong?",
"turn_id": 2
},
{
"speaker": "user",
"text": "My mother-in-law just went into the hospital i... | DailyDialog_test_000042 | home | life_mt_sct_DailyDialog_test_000042_target4 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"social_relationsh... | [
"medical_or_health_context",
"sensitive_content",
"weak_slot_evidence"
] | 0.749 | DailyDialog | test_000042 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | prefix_to_next_assistant_text |
Booking was successful, the total fee is 92.96 GBP payable at the station. Reference number is: 3EPMFSFA. | The visible history is a train booking task in travel transportation. Known constraints are date_or_duration=saturday, sunday; destination=london; transport_type=train. The answer should address the booking or reservation step while tracking departure, destination, and time constraints. | [
{
"speaker": "user",
"text": "Are there any restaurants that serve christmas food around?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I'm sorry, I didn't get any matches for Christmas food. Is there another food type you would like to eat?",
"turn_id": 2
},
{
"speaker": ... | MultiWOZ_PMUL0858.json | travel | life_mt_sct_MultiWOZ_PMUL0858.json_target14 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 1 | MultiWOZ | PMUL0858.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | prefix_to_next_assistant_text |
You need a babysitter. | This sample contains safety-sensitive context. The reasoning should stay conservative, avoid amplifying sensitive details, and only support a safe, bounded assistant response. | [
{
"speaker": "user",
"text": "Listen, Karen, I need your help. I don't know anyone here yet.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I'm glad to help you. What's wrong?",
"turn_id": 2
},
{
"speaker": "user",
"text": "My mother-in-law just went into the hospital i... | DailyDialog_test_000042 | home | life_mt_sct_DailyDialog_test_000042_target6 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"social_relationsh... | [
"low_information_answer",
"medical_or_health_context",
"sensitive_content",
"weak_slot_evidence"
] | 0.74 | DailyDialog | test_000042 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | prefix_to_next_assistant_text |
No thank you that will be all | The visible history is a train booking task in travel transportation. Known constraints are date_or_duration=saturday, sunday; destination=london; transport_type=train. The answer should address the booking or reservation step while tracking departure, destination, and time constraints. | [
{
"speaker": "user",
"text": "Are there any restaurants that serve christmas food around?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I'm sorry, I didn't get any matches for Christmas food. Is there another food type you would like to eat?",
"turn_id": 2
},
{
"speaker": ... | MultiWOZ_PMUL0858.json | travel | life_mt_sct_MultiWOZ_PMUL0858.json_target16 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 0.899 | MultiWOZ | PMUL0858.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | prefix_to_next_assistant_text |
I could have Suzy stay at my place. | This sample contains safety-sensitive context. The reasoning should stay conservative, avoid amplifying sensitive details, and only support a safe, bounded assistant response. | [
{
"speaker": "user",
"text": "Listen, Karen, I need your help. I don't know anyone here yet.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I'm glad to help you. What's wrong?",
"turn_id": 2
},
{
"speaker": "user",
"text": "My mother-in-law just went into the hospital i... | DailyDialog_test_000042 | home | life_mt_sct_DailyDialog_test_000042_target8 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"social_relationsh... | [
"medical_or_health_context",
"sensitive_content",
"weak_slot_evidence"
] | 0.749 | DailyDialog | test_000042 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | prefix_to_next_assistant_text |
Why not? | This sample contains safety-sensitive context. The reasoning should stay conservative, avoid amplifying sensitive details, and only support a safe, bounded assistant response. | [
{
"speaker": "user",
"text": "Listen, Karen, I need your help. I don't know anyone here yet.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I'm glad to help you. What's wrong?",
"turn_id": 2
},
{
"speaker": "user",
"text": "My mother-in-law just went into the hospital i... | DailyDialog_test_000042 | home | life_mt_sct_DailyDialog_test_000042_target10 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"social_relationsh... | [
"medical_or_health_context",
"sensitive_content",
"too_short_answer",
"weak_slot_evidence",
"weak_target_answer"
] | 0.5 | DailyDialog | test_000042 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | prefix_to_next_assistant_text |
I know who we can ask. There is a girl I know, Sara Ralston.She is 17, and she will usually take this kind of job. I know her from my church, and I trust her completely. | This sample contains safety-sensitive context. The reasoning should stay conservative, avoid amplifying sensitive details, and only support a safe, bounded assistant response. | [
{
"speaker": "user",
"text": "Listen, Karen, I need your help. I don't know anyone here yet.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I'm glad to help you. What's wrong?",
"turn_id": 2
},
{
"speaker": "user",
"text": "My mother-in-law just went into the hospital i... | DailyDialog_test_000042 | home | life_mt_sct_DailyDialog_test_000042_target12 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"social_relationsh... | [
"medical_or_health_context",
"sensitive_content",
"weak_slot_evidence"
] | 0.749 | DailyDialog | test_000042 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | prefix_to_next_assistant_text |
I'll get you her number. I hope she can do it. She is very responsible. | This sample contains safety-sensitive context. The reasoning should stay conservative, avoid amplifying sensitive details, and only support a safe, bounded assistant response. | [
{
"speaker": "user",
"text": "Listen, Karen, I need your help. I don't know anyone here yet.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I'm glad to help you. What's wrong?",
"turn_id": 2
},
{
"speaker": "user",
"text": "My mother-in-law just went into the hospital i... | DailyDialog_test_000042 | home | life_mt_sct_DailyDialog_test_000042_target14 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"social_relationsh... | [
"medical_or_health_context",
"sensitive_content",
"weak_slot_evidence"
] | 0.749 | DailyDialog | test_000042 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | prefix_to_next_assistant_text |
I'll go home and get Sara's number. I'll call her first myself. | This sample contains safety-sensitive context. The reasoning should stay conservative, avoid amplifying sensitive details, and only support a safe, bounded assistant response. | [
{
"speaker": "user",
"text": "Listen, Karen, I need your help. I don't know anyone here yet.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I'm glad to help you. What's wrong?",
"turn_id": 2
},
{
"speaker": "user",
"text": "My mother-in-law just went into the hospital i... | DailyDialog_test_000042 | home | life_mt_sct_DailyDialog_test_000042_target16 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"social_relationsh... | [
"medical_or_health_context",
"sensitive_content",
"weak_slot_evidence"
] | 0.749 | DailyDialog | test_000042 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the visible dialogue prefix is used. The turn is handled conservatively because the context is safety-sensitive, so the state keeps prior safe context and waits for a safe next response.",
"reasoning_confidence": "medium",
"reasoning_type": "safet... | prefix_to_next_assistant_text |
A draft beer please. | The dialogue is uncertain context in general daily life, with information request intent. Key constraints are deadline=evening. The answer is grounded because it responds to that goal using only those stated details. | [
{
"speaker": "user",
"text": "Good evening. What'll you have?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I'd like a beer. What kind of beer do you have?",
"turn_id": 2
},
{
"speaker": "user",
"text": "Draft beer and cans beer. Which do you prefer?",
"turn_id": 3... | DailyDialog_test_000043 | study | life_mt_sct_DailyDialog_test_000043_target4 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"general_daily_life"
],
"domain_catego... | [
"low_information_answer",
"low_taxonomy_confidence"
] | 0.749 | DailyDialog | test_000043 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: deadline; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: deadline; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
Could I have some fruit? | The dialogue is uncertain context in general daily life, with confirmation clarification intent. Key constraints are deadline=evening. The answer is grounded because it responds to that goal using only those stated details. | [
{
"speaker": "user",
"text": "Good evening. What'll you have?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I'd like a beer. What kind of beer do you have?",
"turn_id": 2
},
{
"speaker": "user",
"text": "Draft beer and cans beer. Which do you prefer?",
"turn_id": 3... | DailyDialog_test_000043 | study | life_mt_sct_DailyDialog_test_000043_target6 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"general_daily_life"
],
"domain_catego... | [
"low_taxonomy_confidence"
] | 0.749 | DailyDialog | test_000043 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: deadline; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: deadline; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
What are those dates? | The visible history is a flight booking task in travel transportation. Known constraints are date_or_duration=afternoon; destination=san francisco; transport_type=flight. The answer should address the booking or reservation step while tracking origin, destination, and travel dates. | [
{
"speaker": "user",
"text": "Hello, I need a round trip flight to San Francisco.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Hello, how can I help you? Can you give me some specifications?",
"turn_id": 2
},
{
"speaker": "user",
"text": "Yes. I need to be sure if it ... | Taskmaster_dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c | travel | life_mt_sct_Taskmaster_dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c_target4 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [] | 0.899 | Taskmaster | dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoning_confi... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoning_confi... | prefix_to_next_assistant_text |
Yes, a very pleasant journey indeed. Thank you. | The dialogue is travel planning in travel transportation, with emotional support intent. Key constraints are destination=new york; origin=new york. The answer is grounded because it responds to that goal using only those stated details. | [
{
"speaker": "user",
"text": "Hello, Mr. Black, how are you?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Fine, thank you, and how are you?",
"turn_id": 2
},
{
"speaker": "user",
"text": "Very well, thank you. It's nice to meet you again. I am awfully sorry for not be... | DailyDialog_test_000044 | travel | life_mt_sct_DailyDialog_test_000044_target4 | true | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"travel_transportation"
],
"domain_cat... | [] | 0.899 | DailyDialog | test_000044 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
Okay, I'd recommend Golden Wok. It's a lovely chinese restaurant at 191 Histon Road Chesterton. Would that work for you? | The visible history is a restaurant search task in food dining. It should track cuisine, area, and price. The answer should ask, confirm, or respond only to details present in the visible history. | [
{
"speaker": "user",
"text": "I'd like to find a place to dine in the north please.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I have several restaurants to recommend in this area. Would you like a particular style of food?",
"turn_id": 2
},
{
"speaker": "user",
"te... | MultiWOZ_MUL1271.json | travel | life_mt_sct_MultiWOZ_MUL1271.json_target4 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [
"weak_final_state"
] | 0.97 | MultiWOZ | MUL1271.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | prefix_to_next_assistant_text |
Do you have a price limit? | The visible history is a flight booking task in travel transportation. Known constraints are date_or_duration=afternoon; destination=san francisco; transport_type=flight. The answer should address the booking or reservation step while tracking origin, destination, and travel dates. | [
{
"speaker": "user",
"text": "Hello, I need a round trip flight to San Francisco.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Hello, how can I help you? Can you give me some specifications?",
"turn_id": 2
},
{
"speaker": "user",
"text": "Yes. I need to be sure if it ... | Taskmaster_dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c | travel | life_mt_sct_Taskmaster_dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c_target6 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [] | 0.899 | Taskmaster | dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoning_confi... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoning_confi... | prefix_to_next_assistant_text |
Not bad. The fur market is not very brisk lately, but the selling season is advancing near. I hope there will be more buyers in the market this year. | The dialogue is travel planning in travel transportation, with information request intent. Key constraints are destination=new york; origin=new york. The answer is grounded because it responds to that goal using only those stated details. | [
{
"speaker": "user",
"text": "Hello, Mr. Black, how are you?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Fine, thank you, and how are you?",
"turn_id": 2
},
{
"speaker": "user",
"text": "Very well, thank you. It's nice to meet you again. I am awfully sorry for not be... | DailyDialog_test_000044 | travel | life_mt_sct_DailyDialog_test_000044_target6 | true | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"travel_transportation"
],
"domain_cat... | [] | 1 | DailyDialog | test_000044 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
I have booked you a table for 3 at the Golden Wok on Friday at 18:30, your reference number is 6D2IMEYU. Can I help you with anything more? | The visible history is a restaurant booking task in food dining. Known constraints are date_or_duration=friday. The answer should address the booking or reservation step while tracking cuisine, area, price, time, and party size. | [
{
"speaker": "user",
"text": "I'd like to find a place to dine in the north please.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I have several restaurants to recommend in this area. Would you like a particular style of food?",
"turn_id": 2
},
{
"speaker": "user",
"te... | MultiWOZ_MUL1271.json | travel | life_mt_sct_MultiWOZ_MUL1271.json_target6 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 1 | MultiWOZ | MUL1271.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | prefix_to_next_assistant_text |
What seating class do you prefer? | The visible history is a flight booking task in travel transportation. Known constraints are budget=7, 21, $1,700; date_or_duration=afternoon; destination=san francisco; transport_type=flight. The answer should address the booking or reservation step while tracking origin, destination, and travel dates. | [
{
"speaker": "user",
"text": "Hello, I need a round trip flight to San Francisco.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Hello, how can I help you? Can you give me some specifications?",
"turn_id": 2
},
{
"speaker": "user",
"text": "Yes. I need to be sure if it ... | Taskmaster_dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c | travel | life_mt_sct_Taskmaster_dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c_target8 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [] | 0.899 | Taskmaster | dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoning_confi... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoning_confi... | prefix_to_next_assistant_text |
You are welcome. | The dialogue is uncertain context in general daily life, with information request intent. The most recent user context is: What? Oh, what we live! Thank you very much for your help, miss. The answer is grounded only in the visible dialogue history and should stay conservative. | [
{
"speaker": "user",
"text": "By the way miss, where is the toilet?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Toilets are in the rear, I am afraid all the toilets are fully occupied at the moment.",
"turn_id": 2
},
{
"speaker": "user",
"text": "What? Oh, what we li... | DailyDialog_test_000045 | daily_advice | life_mt_sct_DailyDialog_test_000045_target4 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"general_daily_life"
],
"domain_catego... | [
"answer_not_grounded",
"fragment_answer",
"generic_answer",
"low_taxonomy_confidence",
"too_short_answer",
"weak_final_state",
"weak_target_answer"
] | 0 | DailyDialog | test_000045 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: none; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",
... | prefix_to_next_assistant_text |
Lovell Lodge moderately priced and is located on 365 Milton Road. Would you like a reservation? | The visible history is a restaurant booking task in food dining. Known constraints are date_or_duration=friday; hotel_preference=hotel. The answer should address the booking or reservation step while tracking cuisine, area, price, time, and party size. | [
{
"speaker": "user",
"text": "I'd like to find a place to dine in the north please.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I have several restaurants to recommend in this area. Would you like a particular style of food?",
"turn_id": 2
},
{
"speaker": "user",
"te... | MultiWOZ_MUL1271.json | travel | life_mt_sct_MultiWOZ_MUL1271.json_target8 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 1 | MultiWOZ | MUL1271.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | prefix_to_next_assistant_text |
I found a flight for you, it's Virgin Airlines. | The visible history is a flight booking task in travel transportation. Known constraints are budget=7, 21, $1,700; date_or_duration=afternoon; destination=san francisco; transport_type=flight. The answer should address the booking or reservation step while tracking origin, destination, and travel dates. | [
{
"speaker": "user",
"text": "Hello, I need a round trip flight to San Francisco.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Hello, how can I help you? Can you give me some specifications?",
"turn_id": 2
},
{
"speaker": "user",
"text": "Yes. I need to be sure if it ... | Taskmaster_dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c | travel | life_mt_sct_Taskmaster_dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c_target10 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [] | 0.899 | Taskmaster | dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoning_confi... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoning_confi... | prefix_to_next_assistant_text |
it takes about 3 minutes per person, so it ' ll probably be about an hour ' s wait. | The dialogue is shopping assistance in shopping retail, with information request intent. Key constraints are item_type=bubble wrap and to post a packa. The answer is grounded because it responds to that goal using only those stated details. | [
{
"speaker": "user",
"text": "excuse me, could you tell me which line I ' m supposed to stand in to buy bubble wrap and to post a package?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "you can buy the bubble wrap here, but you ' ll have to stand in line over here to post your passage.... | DailyDialog_test_000046 | shopping | life_mt_sct_DailyDialog_test_000046_target4 | false | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categor... | [
"answer_not_grounded",
"off_topic_answer"
] | 0.3 | DailyDialog | test_000046 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: item_type; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: item_type; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",... | prefix_to_next_assistant_text |
Your room at the Lovell Lodge is all set (Ref #5HMHIBZ2). Can I help you with anything else? | The visible history is a restaurant booking task in food dining. Known constraints are date_or_duration=friday; hotel_preference=hotel. The answer should address the booking or reservation step while tracking cuisine, area, price, time, and party size. | [
{
"speaker": "user",
"text": "I'd like to find a place to dine in the north please.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I have several restaurants to recommend in this area. Would you like a particular style of food?",
"turn_id": 2
},
{
"speaker": "user",
"te... | MultiWOZ_MUL1271.json | travel | life_mt_sct_MultiWOZ_MUL1271.json_target10 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [] | 1 | MultiWOZ | MUL1271.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | prefix_to_next_assistant_text |
Departs at 5:29 PM and arrives at 8:57 PM. | The visible history is a flight booking task in travel transportation. Known constraints are budget=7, 21, $1,700; date_or_duration=afternoon; destination=san francisco; transport_type=flight. The answer should address the confirmation clarification step while tracking origin, destination, and travel dates. | [
{
"speaker": "user",
"text": "Hello, I need a round trip flight to San Francisco.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Hello, how can I help you? Can you give me some specifications?",
"turn_id": 2
},
{
"speaker": "user",
"text": "Yes. I need to be sure if it ... | Taskmaster_dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c | travel | life_mt_sct_Taskmaster_dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c_target12 | true | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [] | 1 | Taskmaster | dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoning_confi... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoning_confi... | prefix_to_next_assistant_text |
sure. How many would you like? | The dialogue is shopping assistance in shopping retail, with information request intent. Key constraints are item_type=bubble wrap and to post a packa, stamps here. The answer is grounded because it responds to that goal using only those stated details. | [
{
"speaker": "user",
"text": "excuse me, could you tell me which line I ' m supposed to stand in to buy bubble wrap and to post a package?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "you can buy the bubble wrap here, but you ' ll have to stand in line over here to post your passage.... | DailyDialog_test_000046 | shopping | life_mt_sct_DailyDialog_test_000046_target6 | true | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categor... | [] | 0.899 | DailyDialog | test_000046 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: item_type; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: item_type; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",... | prefix_to_next_assistant_text |
You're welcome, enjoy Cambridge! | The visible history is a restaurant booking task in food dining. Known constraints are date_or_duration=friday; hotel_preference=hotel. The answer should address the booking or reservation step while tracking cuisine, area, price, time, and party size. | [
{
"speaker": "user",
"text": "I'd like to find a place to dine in the north please.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "I have several restaurants to recommend in this area. Would you like a particular style of food?",
"turn_id": 2
},
{
"speaker": "user",
"te... | MultiWOZ_MUL1271.json | travel | life_mt_sct_MultiWOZ_MUL1271.json_target12 | false | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categories": [
"food_dini... | [
"low_information_answer"
] | 0.749 | MultiWOZ | MUL1271.json | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible multiwoz prefix is used. The user turn is treated as new task evidence; added slots: none; changed slots: none; removed slots: none; new missing slots: destination; resolved missing slots: none.",
"reasoning_confidence": "high",
... | prefix_to_next_assistant_text |
This airline has free Wi-fi. | The visible history is a flight booking task in travel transportation. Known constraints are budget=7, 21, $1,700; date_or_duration=afternoon; destination=san francisco; transport_type=flight. The answer should address the confirmation clarification step while tracking origin, destination, and travel dates. | [
{
"speaker": "user",
"text": "Hello, I need a round trip flight to San Francisco.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Hello, how can I help you? Can you give me some specifications?",
"turn_id": 2
},
{
"speaker": "user",
"text": "Yes. I need to be sure if it ... | Taskmaster_dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c | travel | life_mt_sct_Taskmaster_dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c_target14 | false | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [
"low_information_answer"
] | 0.749 | Taskmaster | dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoning_confi... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoning_confi... | prefix_to_next_assistant_text |
are you sending them abroad? | The dialogue is shopping assistance in shopping retail, with information request intent. Key constraints are item_type=bubble wrap and to post a packa, stamps here. The answer is grounded because it responds to that goal using only those stated details. | [
{
"speaker": "user",
"text": "excuse me, could you tell me which line I ' m supposed to stand in to buy bubble wrap and to post a package?",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "you can buy the bubble wrap here, but you ' ll have to stand in line over here to post your passage.... | DailyDialog_test_000046 | shopping | life_mt_sct_DailyDialog_test_000046_target8 | true | {
"category": "daily_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain_categor... | [] | 0.899 | DailyDialog | test_000046 | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: item_type; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible dailydialog prefix is used. The user turn updates the daily dialogue context and response policy without relying on future turns; added slots: item_type; changed slots: none; removed slots: none.",
"reasoning_confidence": "medium",... | prefix_to_next_assistant_text |
For first class $2997 round trip. | The visible history is a flight booking task in travel transportation. Known constraints are budget=7, 21, $1,700; date_or_duration=afternoon; destination=san francisco; transport_type=flight. The answer should address the booking or reservation step while tracking origin, destination, and travel dates. | [
{
"speaker": "user",
"text": "Hello, I need a round trip flight to San Francisco.",
"turn_id": 1
},
{
"speaker": "assistant",
"text": "Hello, how can I help you? Can you give me some specifications?",
"turn_id": 2
},
{
"speaker": "user",
"text": "Yes. I need to be sure if it ... | Taskmaster_dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c | travel | life_mt_sct_Taskmaster_dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c_target16 | false | {
"category": "task_oriented_dialogue",
"category_taxonomy_method": "deterministic_source_aware_rules_v1",
"chunk_level": "turn",
"cleaning_flags": [
"too_many_turns",
"too_short_average_turn"
],
"construction_type": "prefix_to_next_assistant_turn",
"deep_reasoning_method": "rule_based",
"domain... | [
"low_information_answer"
] | 0.749 | Taskmaster | dlg-01af92d0-3ba7-42ac-b2d1-3af8ac82df2c | train | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoning_confi... | [
{
"chunk_id": "C1",
"reasoning": "After turn 1, only the current visible taskmaster prefix is used. The user turn is treated as new task evidence; added slots: destination, transport_type; changed slots: none; removed slots: none; new missing slots: none; resolved missing slots: none.",
"reasoning_confi... | prefix_to_next_assistant_text |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.