id stringlengths 64 64 | source stringclasses 3
values | decision_type stringclasses 1
value | status stringclasses 1
value | unsupported_reason stringclasses 0
values | content_hash stringlengths 64 64 | state dict | candidates listlengths 1 70 | target dict | ordered_targets listlengths 0 0 | labels dict | provenance dict | training dict | training_split stringclasses 1
value |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
e9a3a97b3500c0e0c4e4373df9e9335838187afde66aa75e17ac4bc728003348 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | b709f130ed5f14c64db61cd6eeeaee801d825b83657044ffbf37d6e79c3d8a1f | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::check_equipment_availability",
"name": "check_equipment_availability",
"description": "Check availability of agricultural equipment considering user tier limits, maintenance schedules, and existing reservations. Returns availability status with constraints.",
"parameters_json": "{\"pr... | {
"candidate_id": "tool::generate_invoice",
"action_name": "generate_invoice",
"arguments_json": "{\"include_usage_overage\":true,\"reference_id\":\"CONTRACT-2024-7788\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.34375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 320,
"source_identity": "626",
"trajectory_id": "626",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
538c4c724dc48fbda79dba39b6001cbc7b50581e01f94163f67bb396d4c4e8d5 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 5d78c3574fad0d846cc5b615d882a84edb3f0415a1f90b3619accfccfa6f9cb7 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_customer",
"name": "authenticate_customer",
"description": "Authenticate customer using Customer ID and Equipment PIN.\nRequired before any account-sensitive operations.",
"parameters_json": "{\"properties\":{\"customer_id\":{\"description\":\"Unique customer identifier\"... | {
"candidate_id": "tool::get_equipment_details",
"action_name": "get_equipment_details",
"arguments_json": "{\"serial_number\":\"JH-TR2023-4587\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.09375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 322,
"source_identity": "410",
"trajectory_id": "410",
"step_index": 1,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
9f501468c5dd384eecef7ac2bea2aa34bd582cc1ab43a73e55005e0505b524a4 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 389814ef6975c3ab5b78cf48b866c6e692289895c47cab9a7322af7dee675014 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::book_shipment",
"name": "book_shipment",
"description": "Creates a new shipment booking with validated package dimensions, weight restrictions, and customs forms for international shipments. Validates against restricted items and pickup deadlines.",
"parameters_json": "{\"properties\"... | {
"candidate_id": "tool::transfer_to_human_agent",
"action_name": "transfer_to_human_agent",
"arguments_json": "{\"issue_category\":\"legal_docs\",\"summary\":\"User reports needing to ship a recalled Class II medical device subject to FDA safety alert. They are an online seller requesting guidance on proper prot... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.3125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 323,
"source_identity": "214",
"trajectory_id": "214",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
f797898a50d76084499c046d44951d57c61071824a093e67d4382c96e2a2d758 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | b44958b4a55c1b69f6469db8e2ed9c2a694f2fc5c6b26a84892cada330acad5e | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Verify user identity using User ID + registered phone number OR Booking confirmation code + last 4 digits of payment card.",
"parameters_json": "{\"properties\":{\"booking_confirmation\":{\"description\":\"Booking confir... | {
"candidate_id": "tool::check_seating_availability",
"action_name": "check_seating_availability",
"arguments_json": "{\"screening_id\":\"RS-20241222-1\",\"tier\":\"Standard\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.09375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 324,
"source_identity": "8",
"trajectory_id": "8",
"step_index": 3,
"decision_ordinal": 0... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
cf2ad78e394c71ae03a318875d8fd9d20c7d252bf9148b39609903339ed7afa4 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | a80c38f289a8f1f59316389b042660e87495d51ed70791f5614b80c1cd4c0eb6 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Verify user identity using User ID + registered phone number OR Booking confirmation code + last 4 digits of payment card.",
"parameters_json": "{\"properties\":{\"booking_confirmation\":{\"description\":\"Booking confir... | {
"candidate_id": "tool::check_seating_availability",
"action_name": "check_seating_availability",
"arguments_json": "{\"screening_id\":\"RS-20241222-1\",\"tier\":\"Premium\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.21875,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 327,
"source_identity": "8",
"trajectory_id": "8",
"step_index": 4,
"decision_ordinal": 0... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
8c26b17234f3f187061702a2b03deaaa96f7a1c2aa1000cdab89e10ca29d8c7c | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 8a7b56757e0696b3ccbd4a9e67fc3f768e9a3e131f544d000dda71c075db1a2a | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Authenticate user using driver's license ID and reservation confirmation code. Required before accessing reservation details or performing actions.",
"parameters_json": "{\"properties\":{\"driver_license_id\":{\"descript... | {
"candidate_id": "tool::authenticate_user",
"action_name": "authenticate_user",
"arguments_json": "{\"driver_license_id\":\"WA2223334\",\"reservation_confirmation_code\":\"I5K9J1\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.46875,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 329,
"source_identity": "854",
"trajectory_id": "854",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
01eac62391fb0b99d7e445471b1e13577924cca081b730258c25af819c58b72d | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 8e59548a4270ac71009f82940bce7c8c47af0628230dc58bcfc62df7944dc239 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::verify_donor_identity",
"name": "verify_donor_identity",
"description": "Authenticates donor identity using email and last 4 digits of phone number",
"parameters_json": "{\"properties\":{\"donor_id\":{\"description\":\"Unique donor identifier\",\"type\":\"string\"},\"email\":{\"descri... | {
"candidate_id": "tool::get_financial_report",
"action_name": "get_financial_report",
"arguments_json": "{\"campaign_id\":\"\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.25,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 330,
"source_identity": "1061",
"trajectory_id": "1061",
"step_index": 0,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
80897fb9c777a3f673f19d8a8f0d2b5042c6dca568ae22ff7636059ee034efbf | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 700c647b3538e9fe782f1ccab5b0f5c720bae6f7bb01a4e2fdf01d50057ed25b | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_applicant",
"name": "authenticate_applicant",
"description": "Authenticate applicant using either: Member ID + Date of Birth + Last 4 SSN digits OR Full Name + Registered Email + ZIP Code",
"parameters_json": "{\"properties\":{\"credentials\":{\"additionalProperties\":tru... | {
"candidate_id": "tool::submit_document",
"action_name": "submit_document",
"arguments_json": "{\"document_ref\":\"Court Order 2025-11-30\",\"document_type\":\"identity_proof\",\"user_id\":\"PTA-7788\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.59375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 331,
"source_identity": "628",
"trajectory_id": "628",
"step_index": 1,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
83562126d669c4dcf28bb00865dc06435f461699b57a519e43cd97c7d856831b | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | b79564e368614c56d17efd315624b74f182cf02baa11c02febc47a86f10f4ebc | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_by_visitor_id",
"name": "authenticate_by_visitor_id",
"description": "Verify visitor identity using visitor ID and full name.",
"parameters_json": "{\"properties\":{\"full_name\":{\"type\":\"string\"},\"visitor_id\":{\"type\":\"string\"}},\"required\":[\"visitor_id\",\"fu... | {
"candidate_id": "tool::check_fire_restrictions",
"action_name": "check_fire_restrictions",
"arguments_json": "{\"park_id\":\"YELL\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.34375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 332,
"source_identity": "10",
"trajectory_id": "10",
"step_index": 0,
"decision_ordinal":... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
e2e4fd1e89022630b71e29327dc2531b957ed2d0fff8d90704e99d15cd9069f3 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 17428670598e764d21a3784eb4434153d4b958306b024faadda63b2b54658332 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::enroll_in_course",
"name": "enroll_in_course",
"description": "Enrolls a user in a course after verifying eligibility, subscription status, and enrollment window availability",
"parameters_json": "{\"properties\":{\"course_id\":{\"description\":\"Course identifier\",\"type\":\"string\... | {
"candidate_id": "tool::get_user_profile",
"action_name": "get_user_profile",
"arguments_json": "{\"user_id\":\"LH-8865\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.5625,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 333,
"source_identity": "1063",
"trajectory_id": "1063",
"step_index": 0,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
55f45864c4a113a190f9a5145280ba79b068b45490fec9a7bbc229f6905f4c75 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 32250770fc2b6f42f37c3c9c67668cd93806b10e1e15e709adc53cda329dcb93 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Authenticate user using driver's license ID and reservation confirmation code. Required before accessing reservation details or performing actions.",
"parameters_json": "{\"properties\":{\"driver_license_id\":{\"descript... | {
"candidate_id": "tool::transfer_to_human_agents",
"action_name": "transfer_to_human_agents",
"arguments_json": "{\"summary\":\"Customer disputing $1000 deductible applied to $950 damage claim for reservation I5K9J1. Customer claims damage was minor collision below deductible threshold and requests waiver review... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 336,
"source_identity": "854",
"trajectory_id": "854",
"step_index": 2,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
b38447c3d533d2d8762c6110f2d8bbde29fc00cb947716826505f6090139b074 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 42c2f97dc7fc80abc96ba262ab7f12e77e6a699f8f54c1b26a1be490882935af | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_customer",
"name": "authenticate_customer",
"description": "Authenticate customer using Customer ID and Equipment PIN.\nRequired before any account-sensitive operations.",
"parameters_json": "{\"properties\":{\"customer_id\":{\"description\":\"Unique customer identifier\"... | {
"candidate_id": "tool::get_warranty_claim_status",
"action_name": "get_warranty_claim_status",
"arguments_json": "{\"claim_id\":\"CLM-2023-771\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.21875,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 337,
"source_identity": "410",
"trajectory_id": "410",
"step_index": 1,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
d49b466cb4d0c37f246f247d3b50210947178d9781fd4d1507d0538424d59844 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 78d70f83be11df473af2d5e6bf2fa63c7c0273461aac7d3c8a3240de0f061141 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_applicant",
"name": "authenticate_applicant",
"description": "Authenticate applicant using either: Member ID + Date of Birth + Last 4 SSN digits OR Full Name + Registered Email + ZIP Code",
"parameters_json": "{\"properties\":{\"credentials\":{\"additionalProperties\":tru... | {
"candidate_id": "tool::list_certifications",
"action_name": "list_certifications",
"arguments_json": "{\"user_id\":\"PTA-7788\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 338,
"source_identity": "628",
"trajectory_id": "628",
"step_index": 3,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
bdcb69d62025f9bbc2e0fe3695eb2d51a4a1c347c4d82f0e189344d78df71d8f | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 163fc8f92402ace88d091b1c8184253edf665beae97f72e097733d0080dc47ef | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_client",
"name": "authenticate_client",
"description": "Verify client identity using tiered authentication methods. For sensitive actions, full SSN and security answers are required.",
"parameters_json": "{\"properties\":{\"auth_level\":{\"description\":\"Authentication t... | {
"candidate_id": "tool::compare_loan_products",
"action_name": "compare_loan_products",
"arguments_json": "{\"client_id\":\"F8899\",\"down_payment_pct\":5,\"property_value\":350000}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.1875,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 339,
"source_identity": "218",
"trajectory_id": "218",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
5d09ec654b2ed63e21e55707d4d4337bba28da835832d6a1e2ec3cc179b1f897 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 69c5b0eda2f7c65e62bf49cdf267092dc9bb6e2b7d2b0c2add76c84618c4870a | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Authenticate user using company ID and employee email. Required before accessing or modifying company, employee, or enrollment data.",
"parameters_json": "{\"properties\":{\"company_id\":{\"description\":\"Unique company... | {
"candidate_id": "tool::get_trainer_details",
"action_name": "get_trainer_details",
"arguments_json": "{\"trainer_id\":\"FG-TRN-04\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.25,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 341,
"source_identity": "629",
"trajectory_id": "629",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
2af038a2288cff1ff093054f3a30e391994daf778cac1edd5156e64bdffb4a15 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | a670b25b07338e7018e81adcb0ecf8542f255261bb1261cd5498e8775bc07061 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_client",
"name": "authenticate_client",
"description": "Verify client identity using tiered authentication methods. For sensitive actions, full SSN and security answers are required.",
"parameters_json": "{\"properties\":{\"auth_level\":{\"description\":\"Authentication t... | {
"candidate_id": "tool::get_fee_schedule",
"action_name": "get_fee_schedule",
"arguments_json": "{\"fee_type\":\"origination\",\"loan_amount\":332500}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.5,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 342,
"source_identity": "218",
"trajectory_id": "218",
"step_index": 1,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
e33d64af6bf323de73c48db791a41452b792f69bba6960385e28ef51c57933a3 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | c4b8bf2576205ba9d169c185d020404e3b2e1217cd2fa6517208c8f76e413931 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_client",
"name": "authenticate_client",
"description": "Authenticate client for profile access using Client ID and Property ZIP Code.",
"parameters_json": "{\"properties\":{\"client_id\":{\"description\":\"Unique client identifier\",\"type\":\"string\"},\"zip_code\":{\"de... | {
"candidate_id": "tool::transfer_to_human_agent",
"action_name": "transfer_to_human_agent",
"arguments_json": "{\"summary\":\"Client R-8881, Contract S-112, requesting guidance on floating floor system for vibration isolation and specialized foundation work for acoustic integrity. This involves structural engine... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.3125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 345,
"source_identity": "411",
"trajectory_id": "411",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
5e8ac57ff185c900664daded867c8ad8c7220175695e1387ab2b0ea8a8d5ea7a | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 55596d8573e0f776a82dc1fa405e2075060131736f485e3aaf3ba48ca3c5ed4b | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_client",
"name": "authenticate_client",
"description": "Verify client identity using tiered authentication methods. For sensitive actions, full SSN and security answers are required.",
"parameters_json": "{\"properties\":{\"auth_level\":{\"description\":\"Authentication t... | {
"candidate_id": "tool::get_fee_schedule",
"action_name": "get_fee_schedule",
"arguments_json": "{\"fee_type\":\"appraisal\",\"loan_amount\":332500}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.0625,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 346,
"source_identity": "218",
"trajectory_id": "218",
"step_index": 2,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
df8919353873201b77b4a0b55379e7be42a5eaf0d36595f7db327241edd3340b | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 303aaf518a265c120694d0c4919f1e0701f5a81e2843d31bc331adf04dce4f91 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Verify user identity via user ID + registered phone number. Required for all account actions.",
"parameters_json": "{\"properties\":{\"phone_number\":{\"description\":\"Registered phone number including country code\",\"... | {
"candidate_id": "tool::get_user_profile",
"action_name": "get_user_profile",
"arguments_json": "{\"user_id\":\"S776\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.34375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 347,
"source_identity": "412",
"trajectory_id": "412",
"step_index": 2,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
3760c6a25b0b00d9775f7a1ea8dd5cc09b3a9ce7f307326e80e571b35ca5784d | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | dd18fc5f7a47796999de77e71e5ab679571221d551007730f6b7b895036cdc86 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::verify_user_identity",
"name": "verify_user_identity",
"description": "Authenticate users using approved methods: phone+license plate, email+payment card digits, or QR codes",
"parameters_json": "{\"properties\":{\"email\":{\"description\":\"User email address\",\"title\":\"Email\",\"... | {
"candidate_id": "tool::get_user_profile",
"action_name": "get_user_profile",
"arguments_json": "{\"user_id\":\"U123456\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.1875,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 348,
"source_identity": "630",
"trajectory_id": "630",
"step_index": 1,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
cc29cb01c98807444b242e281b2cdec16aa91b48f2e2dbd2f1002e47ad6debf5 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 73c0f9062da1ec1e3232908e7022ff91e084fef9d781eaaa87fb53be7e210af7 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::schedule_appointment",
"name": "schedule_appointment",
"description": "Schedule an assessment, delivery, or pickup appointment. Ensures compliance with time constraints and verifies identity.",
"parameters_json": "{\"properties\":{\"appointment_type\":{\"description\":\"Type of appoin... | {
"candidate_id": "tool::request_quote",
"action_name": "request_quote",
"arguments_json": "{\"artifact_details\":{\"century\":19,\"damage_description\":\"Scratches on the upper right panel, missing inlay in the lower left drawer, and tarnished brass hardware.\",\"description\":\"Hand-carved walnut armoire with i... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.34375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 352,
"source_identity": "1064",
"trajectory_id": "1064",
"step_index": 0,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
515ae4f2d8fa8f696489b714a38f7a59cf0e5f018ad2a0dd95ad603182e2d668 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | b076d2c3cd0e468dcd9b9000865f7b911f3979d5492de32c3a1bf020fe8aeb54 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::schedule_appointment",
"name": "schedule_appointment",
"description": "Schedule an assessment, delivery, or pickup appointment. Ensures compliance with time constraints and verifies identity.",
"parameters_json": "{\"properties\":{\"appointment_type\":{\"description\":\"Type of appoin... | {
"candidate_id": "tool::schedule_appointment",
"action_name": "schedule_appointment",
"arguments_json": "{\"appointment_type\":\"assessment\",\"artifact_id\":\"ART-7890\",\"customer_id\":\"123456\",\"slot_datetime\":\"2023-10-31T10:00:00Z\",\"verification_data\":{\"artifact_registration_number\":\"ART-7890\",\"c... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.53125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 353,
"source_identity": "1064",
"trajectory_id": "1064",
"step_index": 0,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
8d9f4f82c0afc65b16a7fb55b6c4b266ec0acbcc3c9e55c12670ae00b572ef10 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | d8b1e05811512b229c21b2c2a4ddd81d887ae966c3ea880fae1044320bd73d81 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_owner",
"name": "authenticate_owner",
"description": "Verify pet owner identity using:\n\n- Owner ID + Pet ID + ZIP Code, OR\n- Owner Name + Pet Name + Last 4 Phone Digits",
"parameters_json": "{\"properties\":{\"credentials\":{\"additionalProperties\":false,\"description... | {
"candidate_id": "tool::check_vet_availability",
"action_name": "check_vet_availability",
"arguments_json": "{\"appointment_type\":\"urgent\",\"exotic_specialist\":false,\"time_window\":\"today\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.09375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 354,
"source_identity": "14",
"trajectory_id": "14",
"step_index": 1,
"decision_ordinal":... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
ea4526f8178ced5167ffe98886a95dd5696244fc0f012d42610ad24a307f5034 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 5e15a531e7cf45f19160751f613c04f26024d9cf987b65fe6734d1677d6f4a61 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::schedule_exam",
"name": "schedule_exam",
"description": "Schedule an eye exam appointment",
"parameters_json": "{\"properties\":{\"appointment_type\":{\"enum\":[\"comprehensive\",\"contact_fitting\",\"specialty\"],\"type\":\"string\"},\"preferred_date\":{\"format\":\"date\",\"type\":\... | {
"candidate_id": "tool::get_order_status",
"action_name": "get_order_status",
"arguments_json": "{\"order_id\":\"O55667\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.53125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 355,
"source_identity": "219",
"trajectory_id": "219",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
166fe92adb9fe4d6070b95771e30f883a31a7ba9bbb25ea25570b69543164f50 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | a88f8f9f15fea2d40b6a593d0b38186c97860c46104c7fc9a8e2e61790923a6a | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Verify user identity via email/username and SMS verification code. Required before account-sensitive actions.",
"parameters_json": "{\"properties\":{\"sms_code\":{\"type\":\"string\"},\"username_or_email\":{\"type\":\"st... | {
"candidate_id": "tool::get_user_profile",
"action_name": "get_user_profile",
"arguments_json": "{\"user_id\":\"D2-USER-888\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.5,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 356,
"source_identity": "857",
"trajectory_id": "857",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
eb7cc8bcdeff3cd95e55936a5f8b16a93003922cae9ab0bda22595c46fb49969 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 43915afeec6d6be44f8611d44faa25d63b5b20f4c2d8f25ed05d58cd4591e9dd | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_client",
"name": "authenticate_client",
"description": "Verify client identity using tiered authentication methods. For sensitive actions, full SSN and security answers are required.",
"parameters_json": "{\"properties\":{\"auth_level\":{\"description\":\"Authentication t... | {
"candidate_id": "tool::transfer_to_loan_officer",
"action_name": "transfer_to_loan_officer",
"arguments_json": "{\"summary\":\"Client seeking guidance on first-time buyer program eligibility for spouse (Client ID F8899). Client has previous homeownership history, spouse has 740 credit score. Questions about dow... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.5,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 357,
"source_identity": "218",
"trajectory_id": "218",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
af05496d37a6dc9f0e026cb652e22cab786f8d4e6033775eff8552a0c75b2ab9 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 129b488fd55cb2db30877d5f9ca42389262b4d4cb1f27143de188964b169230a | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_client",
"name": "authenticate_client",
"description": "Verify client identity for account holders or guests. For members: name + contact. For guests: name + booking_confirmation.",
"parameters_json": "{\"properties\":{\"credentials\":{\"properties\":{\"booking_confirmati... | {
"candidate_id": "tool::transfer_to_human_agent",
"action_name": "transfer_to_human_agent",
"arguments_json": "{\"issue_category\":\"dispute\",\"summary\":\"Client Sarah Johnson (CL-948273) disputing no-show fee for booking B-729184. Client arrived 3 hours late (10 AM PST instead of 10 AM EST) claiming booking s... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 358,
"source_identity": "414",
"trajectory_id": "414",
"step_index": 2,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
55dd394f926a1cd47b9d40b6bb359a036cafba6cbd7adb4919c8cdf003ee6310 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | a86060ff889f941cecd715039e04b6c7141ea8f3cefa4434b36f04b327c4e4fd | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::verify_user_identity",
"name": "verify_user_identity",
"description": "Authenticate registered users using username and email/SMS verification",
"parameters_json": "{\"properties\":{\"code\":{\"description\":\"Verification code\",\"type\":\"string\"},\"username\":{\"description\":\"Us... | {
"candidate_id": "tool::get_refund_history",
"action_name": "get_refund_history",
"arguments_json": "{\"user_id\":\"a1b2c3d4-e5f6-7890-g1h2-i3j4k5l6m7n8\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 360,
"source_identity": "632",
"trajectory_id": "632",
"step_index": 1,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
28179deec202831fd41e115e8642b0e32521642048c566a80cf3b229ba6ad643 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | cbd15a2391f2b70edfd2d2441f296b17e2af4621fcfe3155b70bc1b8466f7e72 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_customer",
"name": "authenticate_customer",
"description": "Authenticates customer using customer ID and shipment reference number. Must be completed before any shipment-related actions.",
"parameters_json": "{\"properties\":{\"customer_id\":{\"description\":\"Unique cust... | {
"candidate_id": "tool::get_shipment_details",
"action_name": "get_shipment_details",
"arguments_json": "{\"shipment_reference\":\"SHP-2025-PACK\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.59375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 361,
"source_identity": "416",
"trajectory_id": "416",
"step_index": 1,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
585c3b1dd76898753252a0b1d6bfb6f3e558277c3217df1f02d5071ae57d8222 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | d2ab59b623362cd5d66825b7e8a815b466470e7c9d486e6621f75f8760e4bd6e | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::get_user_details",
"name": "get_user_details",
"description": "Retrieves user profile information including membership tier and payment methods.",
"parameters_json": "{\"properties\":{\"user_id\":{\"description\":\"Unique user identifier\",\"type\":\"string\"}},\"required\":[\"user_id... | {
"candidate_id": "tool::get_user_details",
"action_name": "get_user_details",
"arguments_json": "{\"user_id\":\"SP443322\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.59375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 362,
"source_identity": "1066",
"trajectory_id": "1066",
"step_index": 1,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
e1b7505b953bce917f463d8cf7197dbaac1ffdb1227a1c498d7739a954788a50 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | b06b3c7b1b79e713b63b6f09a6938bc232aac5a884dfa11a606e5701998e4eb2 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::schedule_appointment",
"name": "schedule_appointment",
"description": "Book a new auto repair appointment with loyalty and emergency slot validation",
"parameters_json": "{\"properties\":{\"customer_id\":{\"description\":\"Unique customer identifier\",\"type\":\"string\"},\"is_emergen... | {
"candidate_id": "tool::get_customer_profile",
"action_name": "get_customer_profile",
"arguments_json": "{\"customer_id\":\"F123456\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 363,
"source_identity": "1067",
"trajectory_id": "1067",
"step_index": 0,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
3a517b0062d9ae7c46f459cded6a3901d24383d8b9a858dd251277ff3ada0a1b | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 7bc91a90ce8b8baa7bbc8d400e81b5aeccaaa41d2e62cf314cdfebe3483a817d | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Verify user identity via email/username and SMS verification code. Required before account-sensitive actions.",
"parameters_json": "{\"properties\":{\"sms_code\":{\"type\":\"string\"},\"username_or_email\":{\"type\":\"st... | {
"candidate_id": "tool::transfer_to_human_agent",
"action_name": "transfer_to_human_agent",
"arguments_json": "{\"issue_summary\":\"User D2-USER-888 (username: CasualDotaFan88) cannot receive SMS verification codes needed for identity authentication to resolve account flag status. User confirms registered phone ... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.25,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 364,
"source_identity": "857",
"trajectory_id": "857",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
9e6d67ff3df034d525e93caec4daa7accf2fbf6d61fbac6a51007195c5c5e377 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | e63b76106d47f16b4190beb8295f6348dbd109885e88843d9d04466b81b35a29 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Authenticate user using either:\n\n- User ID + Registered Email + Last 4 Phone Digits, OR\n- Order ID + Billing ZIP Code",
"parameters_json": "{\"properties\":{\"auth_method\":{\"description\":\"Authentication method typ... | {
"candidate_id": "tool::authenticate_user",
"action_name": "authenticate_user",
"arguments_json": "{\"auth_method\":\"order_credentials\",\"credentials\":{\"billing_zip\":\"90210\",\"order_id\":\"789456123\"}}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.25,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 365,
"source_identity": "17",
"trajectory_id": "17",
"step_index": 0,
"decision_ordinal":... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
cfb41aa2de2581879dbd1a0fecc5ecbf038d32eeeeeea804f087c91b45a27ce9 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | e2b247def12b2cf5ccb66dbffa720d00fb4483f1e0a6a5e1431890c6cc803c18 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::schedule_appointment",
"name": "schedule_appointment",
"description": "Book a new auto repair appointment with loyalty and emergency slot validation",
"parameters_json": "{\"properties\":{\"customer_id\":{\"description\":\"Unique customer identifier\",\"type\":\"string\"},\"is_emergen... | {
"candidate_id": "tool::get_repair_status",
"action_name": "get_repair_status",
"arguments_json": "{\"repair_id\":\"REPAIR-2023-045678901235\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 366,
"source_identity": "1067",
"trajectory_id": "1067",
"step_index": 1,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
5f53a1365ae5bd4a93802acb08adc0ca129a1782c4dfaba8669495315ef88967 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | cf994d401463df05221764b2c90b74a985d435b34cd471f0508f3514fe1c6ce1 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::check_attraction_status",
"name": "check_attraction_status",
"description": "Retrieve current operational status and details of a theme park attraction",
"parameters_json": "{\"properties\":{\"attractionId\":{\"description\":\"Unique identifier for the attraction\",\"type\":\"string\"... | {
"candidate_id": "tool::check_attraction_status",
"action_name": "check_attraction_status",
"arguments_json": "{\"attractionId\":\"Galaxy Blaster\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.25,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 367,
"source_identity": "222",
"trajectory_id": "222",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
f5de1183e05deb444cf1a215e6d37702e57972b22c0b17ae9dbd033cf73470ee | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 2c2764ef3c3e89ba777e48e1d70d960a24ee283770060e71525c168f5949ec7d | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::reserve_meeting_room",
"name": "reserve_meeting_room",
"description": "Reserve meeting room after validating availability and subscription limits",
"parameters_json": "{\"properties\":{\"duration\":{\"description\":\"Duration in minutes\",\"type\":\"integer\"},\"room_id\":{\"descripti... | {
"candidate_id": "tool::transfer_to_human_agent",
"action_name": "transfer_to_human_agent",
"arguments_json": "{\"summary\":\"User FreelanceWriter2023 on Starter tier wants to change account-wide mail handling preferences from forwarding to scanning for all future mail. Currently has 2 forwarding addresses activ... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.1875,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 369,
"source_identity": "858",
"trajectory_id": "858",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
2547ccdc53583f5993b68e3a05d8058d0c15d7cb2c61b32df93719125fd4fcf1 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 9d4287e8ea481902923f1f6068ad145c8248f77c32fde5bc10faf5a913a27eeb | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::check_attraction_status",
"name": "check_attraction_status",
"description": "Retrieve current operational status and details of a theme park attraction",
"parameters_json": "{\"properties\":{\"attractionId\":{\"description\":\"Unique identifier for the attraction\",\"type\":\"string\"... | {
"candidate_id": "tool::process_compensation",
"action_name": "process_compensation",
"arguments_json": "{\"incidentType\":\"ride_malfunction\",\"membershipTier\":\"regular\",\"queueMinutes\":45}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.1875,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 370,
"source_identity": "18",
"trajectory_id": "18",
"step_index": 1,
"decision_ordinal":... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
79cb3e17b5086de5a5afad4002ac122e4d09ce1e41a78837bb057a6a8852796d | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 432100ac8e61028eef206c14d4eb5c94350c2d1634ce96858f07cc067cec5427 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Verify user identity via ID+DOB or Name+Phone+ZIP as per policy. Required before accessing PII/PHI.",
"parameters_json": "{\"properties\":{\"customer_id\":{\"description\":\"Customer ID (required for id_dob method)\",\"t... | {
"candidate_id": "tool::get_order_details",
"action_name": "get_order_details",
"arguments_json": "{\"order_id\":\"ORDER-654321\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.59375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 371,
"source_identity": "1068",
"trajectory_id": "1068",
"step_index": 3,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
f8354d23d4b6e9bd5ec5a85e82739fa4cfd97bc3298298b2471d9d203fadb82b | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | fc53205fc91145c8c824c33a7177bc91f78f3b0d985d8e366ce7ccec8e606638 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::get_user_details",
"name": "get_user_details",
"description": "Retrieve user profile including membership tier, payment methods, and favorite players.",
"parameters_json": "{\"properties\":{\"auth_method\":{\"enum\":[\"email\",\"phone\",\"payment_last4\"],\"type\":\"string\"},\"auth_v... | {
"candidate_id": "tool::get_meet_and_greet_slots",
"action_name": "get_meet_and_greet_slots",
"arguments_json": "{\"date_range\":\"2025-07-20 to 2025-07-22\",\"player_id\":\"ShadowReaper\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.25,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 372,
"source_identity": "633",
"trajectory_id": "633",
"step_index": 1,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
b6c4899dc9913af5bd2244b9b7af49b9cd188113cba42e15759517f8ed662e28 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | b3097303b2d400ebe588259d61bb53f43dbebb0d962544e1f8637fa570b07583 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Authenticate user based on role (participant, site staff, investigator) using specific credentials.",
"parameters_json": "{\"properties\":{\"credentials\":{\"properties\":{\"investigator\":{\"properties\":{\"investigator... | {
"candidate_id": "tool::get_participant_status",
"action_name": "get_participant_status",
"arguments_json": "{\"participant_id\":\"PT-5511\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.53125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 373,
"source_identity": "225",
"trajectory_id": "225",
"step_index": 1,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
656ae1a0b4ebeb6ae8cb2108b3f1e33a1bb1ed517b06f0878611a787d99f571e | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 3666502293d44db661d06793dcdc4250cb6fce048b820b02268b1d5a89baf471 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::check_facility_availability",
"name": "check_facility_availability",
"description": "Verifies available time slots for a specific park facility. Checks against blackout dates, existing reservations, and booking constraints (3-365 days advance; 30-day minimum for holidays). Returns availab... | {
"candidate_id": "tool::transfer_to_human_agent",
"action_name": "transfer_to_human_agent",
"arguments_json": "{\"summary\":\"User is a fitness coach requesting contact information for park services department and access to trail maps/surface details for jogging trails to tailor low-impact running routines for c... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.28125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 374,
"source_identity": "224",
"trajectory_id": "224",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
fdd38d540f82bb49ea28f03a4188356e51e33b84fed42e1620eac82bfc1b75ba | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 60db7f6d0e35681721d3d89cf319bfb1765eb9e0369a9be4206e5974bab4fb64 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Verify user identity via ID+DOB or Name+Phone+ZIP as per policy. Required before accessing PII/PHI.",
"parameters_json": "{\"properties\":{\"customer_id\":{\"description\":\"Customer ID (required for id_dob method)\",\"t... | {
"candidate_id": "tool::get_prescription_details",
"action_name": "get_prescription_details",
"arguments_json": "{\"prescription_id\":\"RX-762891\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 376,
"source_identity": "1068",
"trajectory_id": "1068",
"step_index": 0,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
d2f255a5eea8c2ec5d068c1c1acf50d34fe94067d83ec490f2bd6dbdc87b4a67 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 545cac90c79e9ef6b9a3a51e19a9ad967a06ea59580fc9f6a5b43ff88ffd9a29 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_student",
"name": "authenticate_student",
"description": "Authenticates student identity using policy methods: student ID + phone number OR full name + date of birth.",
"parameters_json": "{\"properties\":{\"auth_method\":{\"description\":\"Authentication method type\",\"... | {
"candidate_id": "tool::get_student_profile",
"action_name": "get_student_profile",
"arguments_json": "{\"student_id\":\"S45681\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.5,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 377,
"source_identity": "1069",
"trajectory_id": "1069",
"step_index": 0,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
da776008e1096288c1c21e817690e6e2ee5027a0cfda8f36327cb9f6e81b5e3a | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 7639485c96999a3c70a8814d3f6f6f4cdf0ce4b2789ac26e2a27a68e948b8468 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_guest",
"name": "authenticate_guest",
"description": "Authenticate guest using either:\n- Rental License Number + DOB + State Issuance\nOR\n- Membership ID + Last 4 Digits of Phone Number\nReturns authentication status and guest ID.",
"parameters_json": "{\"properties\":{... | {
"candidate_id": "tool::transfer_to_dockmaster",
"action_name": "transfer_to_dockmaster",
"arguments_json": "{\"summary\":\"Guest inquiring about permission for FPV drone racing event with 12 pilots, requesting 500-meter airspace clearance over waterway. Event organization request outside standard rental service... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.46875,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 378,
"source_identity": "419",
"trajectory_id": "419",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
9c238398c36f72bc3248baa01446a9327236265e40ec5c336dc654be9dda183b | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 56eeb9e469165de7f458b49e8b424e7180c909646fb39aa64e3378beed1b58d3 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::schedule_appointment",
"name": "schedule_appointment",
"description": "Schedule electrical service appointment with NEC compliance checks and hazard validation",
"parameters_json": "{\"properties\":{\"appointment_time\":{\"description\":\"Requested service time in ISO format\",\"type\... | {
"candidate_id": "tool::process_payment",
"action_name": "process_payment",
"arguments_json": "{\"amount\":1350.0,\"payment_method_id\":\"PM-9876543210\",\"payment_type\":\"Deposit\",\"property_address\":{\"city\":\"Tempe\",\"street\":\"963 Education Blvd\",\"zip\":\"85281\"},\"quote_id\":\"QUOTE-789456\",\"user... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 381,
"source_identity": "20",
"trajectory_id": "20",
"step_index": 0,
"decision_ordinal":... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
49a7bbcf045a922d2f7f9969beded26844fc6de57eae1fd347399666cabbe954 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 29b2b78c766145b72e63739c21941009550ad5c2828ff6f2ea85d349390bf610 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::verify_customer_by_id_phone",
"name": "verify_customer_by_id_phone",
"description": "Verify customer identity using customer ID and registered phone number.",
"parameters_json": "{\"properties\":{\"customer_id\":{\"description\":\"Unique customer identifier\",\"type\":\"string\"},\"ph... | {
"candidate_id": "tool::transfer_to_human_agent",
"action_name": "transfer_to_human_agent",
"arguments_json": "{\"summary\":\"Customer SPONSOR9 requests branded invoices with company logo for past services on Vessel EVENT23 (35ft catamaran). Customer is asking about third-party options for invoice customization ... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.25,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 382,
"source_identity": "226",
"trajectory_id": "226",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
c877800a3ae1118f1516f2ad2e19bc6cf128d18a99ddc857085760910e753dca | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | d5b5d3fb7d8d3863f9fcedc2bf73cec8890d147e10304f0334d072706f5563d8 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_student",
"name": "authenticate_student",
"description": "Authenticates student identity using policy methods: student ID + phone number OR full name + date of birth.",
"parameters_json": "{\"properties\":{\"auth_method\":{\"description\":\"Authentication method type\",\"... | {
"candidate_id": "tool::get_instructor_availability",
"action_name": "get_instructor_availability",
"arguments_json": "{\"date_range\":\"2025-08-28/2025-08-29\",\"instrument\":\"piano\",\"require_prime_time\":false}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.25,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 384,
"source_identity": "1069",
"trajectory_id": "1069",
"step_index": 0,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
1dccfdd2e8d76d11c22eb4d2c0fada4203939339bec56a987e446e215d25aaae | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 04b4b65fd640dd3643b3cc4d53372243f71821b5cadc82f169324be1d968886c | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Authenticate users via account or guest credentials. Account holders provide user_id + license number. Guests provide reservation_id + last 4 digits of captain's license.",
"parameters_json": "{\"properties\":{\"last_4_l... | {
"candidate_id": "tool::transfer_to_human_agent",
"action_name": "transfer_to_human_agent",
"arguments_json": "{\"summary\":\"Customer reports discrepancy in captain's license last 4 digits for reservation CH-2025-1129. Customer states they provided 7732 during booking, but system shows stored digits as 4481. Au... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.09375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 385,
"source_identity": "21",
"trajectory_id": "21",
"step_index": 1,
"decision_ordinal":... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
5fffccb50553499f90880e2fa2e56e4ecf4421e8a4a58e3e135abd35f54b2115 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | f3a4376880f4a2a6c4bfeaf0acee47055cddfde69c03219a383dadf003b18936 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::create_event_draft",
"name": "create_event_draft",
"description": "Create a new event draft with basic details.",
"parameters_json": "{\"properties\":{\"client_id\":{\"description\":\"Unique client identifier\",\"type\":\"string\"},\"estimated_attendees\":{\"description\":\"Initial he... | {
"candidate_id": "tool::review_contract",
"action_name": "review_contract",
"arguments_json": "{\"event_id\":\"ONBOARD2025\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.25,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 387,
"source_identity": "636",
"trajectory_id": "636",
"step_index": 2,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
e195693dc9bea119db82f29cbeab5fce85e0b826bf40527ace274a8948d5b3ab | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | e4a4c9e788ecfae96f346e0b0cfb10f9c59acbe3e651efaca31e933a905b0820 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::get_beer_details",
"name": "get_beer_details",
"description": "Retrieve details and availability status for a specific beer product.",
"parameters_json": "{\"properties\":{\"beer_id\":{\"description\":\"Unique identifier for the beer product\",\"type\":\"string\"}},\"required\":[\"bee... | {
"candidate_id": "tool::list_available_beers",
"action_name": "list_available_beers",
"arguments_json": "{\"style_filter\":\"All\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.34375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 388,
"source_identity": "22",
"trajectory_id": "22",
"step_index": 0,
"decision_ordinal":... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
1297734e5e4a4e40e675dc7aeb4884091f89d27218181dffddb396571e665a71 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 12738b1e6b9a49b6dc97136064fb4ebb673556bc2affab2b5f355f348e6912d9 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Authenticate users via User ID and Reservation ID before accessing booking details. Must complete before any booking-related actions.",
"parameters_json": "{\"properties\":{\"reservation_id\":{\"description\":\"Reservati... | {
"candidate_id": "tool::transfer_to_human_agent",
"action_name": "transfer_to_human_agent",
"arguments_json": "{\"summary\":\"User needs help retrieving their Reservation ID. They provided User ID: FINDRES and contact email: events@company.com, but do not have their Reservation ID. Agent cannot authenticate or a... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 389,
"source_identity": "1070",
"trajectory_id": "1070",
"step_index": 0,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
bc4977d93a5c4c65f3a8120ba74ae23e4d838540762db8fd427203bfca5397a4 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | f055bb16e917bae7c398c0a6d2d97e128cce31d783ff912d802d332a13acbab7 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_guest",
"name": "authenticate_guest",
"description": "Authenticate guest identity using reservation code + last name OR loyalty ID + phone number.",
"parameters_json": "{\"properties\":{\"last_name\":{\"description\":\"Guest's last name\",\"type\":\"string\"},\"loyalty_id... | {
"candidate_id": "tool::transfer_to_human_agents",
"action_name": "transfer_to_human_agents",
"arguments_json": "{\"summary\":\"Guest authenticated with reservation RES394041 (Smith) requests to update billing address and ensure tax ID is included on invoice. No tool available for billing/invoice modifications.\... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.1875,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 390,
"source_identity": "423",
"trajectory_id": "423",
"step_index": 2,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
e0d02bc1ee4313423a09e0d7c2ca0eb6aa17535621431c5bb76c512241d31464 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 7905045b27f15f6fdb535ee6ef23ddd3bc9c11dd9d44b699a8151c8826f4b3ee | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Verify user identity via ID/phone/email for account holders or booking ID + last 4 payment digits for guests",
"parameters_json": "{\"properties\":{\"auth_method\":{\"enum\":[\"email_code\",\"sms_code\",\"guest_booking\"... | {
"candidate_id": "tool::get_current_time",
"action_name": "get_current_time",
"arguments_json": "{}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.15625,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 394,
"source_identity": "864",
"trajectory_id": "864",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
d69a883bd15a8ba12abe87436b0c21bac3a836c1f6a913dd64f17a563e79f852 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 4b71e05eff37c96ea373cd77914436aaa8b712330060ea4ac9d0572055420269 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::create_event_draft",
"name": "create_event_draft",
"description": "Create a new event draft with basic details.",
"parameters_json": "{\"properties\":{\"client_id\":{\"description\":\"Unique client identifier\",\"type\":\"string\"},\"estimated_attendees\":{\"description\":\"Initial he... | {
"candidate_id": "tool::modify_booking",
"action_name": "modify_booking",
"arguments_json": "{\"event_id\":\"ONBOARD2025\",\"new_end_time\":\"2025-07-10T17:00:00-07:00\",\"new_start_time\":\"2025-07-10T09:00:00-07:00\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.4375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 395,
"source_identity": "636",
"trajectory_id": "636",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
1af6101468132e14da531d6dc3e8a9b3a03b8d0087f5ddec4eed608d602c7b83 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 481edacc4e8c390ea54f83b8669f3dcf03aff193eaef4f55d88b63c54266f1bd | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_traveler",
"name": "authenticate_traveler",
"description": "Verify traveler identity using primary or secondary authentication methods",
"parameters_json": "{\"properties\":{\"booking_id\":{\"description\":\"Primary booking identifier\",\"type\":\"string\"},\"last_name\":... | {
"candidate_id": "tool::get_traveler_profile",
"action_name": "get_traveler_profile",
"arguments_json": "{\"access_level\":\"full\",\"traveler_id\":\"T187263945\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.53125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 397,
"source_identity": "638",
"trajectory_id": "638",
"step_index": 1,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
499a47008133134644cce49afc07dae3884a5f9558eaaed5956c5288eaf5b8f0 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 3cafe65e31692364b589c9277e86276a847f0191a145cf65bdda9da86b8bd570 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Verify user identity using:\n- Account ID + registered email\nOR\n- Order number + last 4 digits of payment method",
"parameters_json": "{\"properties\":{\"account_id\":{\"description\":\"User account ID (if using accoun... | {
"candidate_id": "tool::get_user_profile",
"action_name": "get_user_profile",
"arguments_json": "{\"user_id\":\"U123456\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.34375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 398,
"source_identity": "25",
"trajectory_id": "25",
"step_index": 2,
"decision_ordinal":... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
5b78638dabd53e2c6e4b7401dddb48a27319d1ccd31e5abb00f6599abe196ecc | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | a71c68202ba8d912ec7e38f5ee60523f51e6cd02093eb3061bada0a3494017a2 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Authenticate user using full name with either email or phone plus security code. Required for profile-level actions.",
"parameters_json": "{\"properties\":{\"email\":{\"description\":\"User's registered email address\",\... | {
"candidate_id": "tool::get_vendor_details",
"action_name": "get_vendor_details",
"arguments_json": "{\"vendor_id\":\"VENUE-2023\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.46875,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 399,
"source_identity": "1072",
"trajectory_id": "1072",
"step_index": 2,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
7977631ba0ae8f331d6fee9f88bc0d810ce85636232a47de2c488ef51b5939b7 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 6252f5330c347614449fb2a388d97abecff7ac02eb9772b018a1cb247e1f544b | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Verify user identity using email or phone number + security PIN",
"parameters_json": "{\"properties\":{\"email_or_phone\":{\"type\":\"string\"},\"security_pin\":{\"type\":\"string\"}},\"required\":[\"email_or_phone\",\"s... | {
"candidate_id": "tool::transfer_to_human_support",
"action_name": "transfer_to_human_support",
"arguments_json": "{\"summary\":\"User requests exception to 24-hour workshop cancellation policy due to medical emergency. Workshop wkshp-6d7e8f scheduled in 22 hours. User has documentation to share.\"}",
"label_j... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.28125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 400,
"source_identity": "24",
"trajectory_id": "24",
"step_index": 0,
"decision_ordinal":... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
082d60447b6cb1839be17932fbe67864b516cd4b3478c83ded52fbe36dcd84ae | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 4942f88dda4009675d4cb8f30452799047f62844561c0db359ada6b16e4c78da | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::search_events",
"name": "search_events",
"description": "Search available events based on date, venue, or artist. Filters events occurring after the current time.",
"parameters_json": "{\"properties\":{\"artist\":{\"description\":\"Performer or production name\",\"title\":\"Artist\",\... | {
"candidate_id": "tool::check_seat_availability",
"action_name": "check_seat_availability",
"arguments_json": "{\"event_id\":\"DANCE2025-08\",\"tier\":\"2\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.25,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 402,
"source_identity": "228",
"trajectory_id": "228",
"step_index": 2,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
5ae9722d314eac75d44ecb8eb3794a52e317c6b9235a70b81b603a1b4e57c200 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 8d03dc41b7de07ec18731877a650d43a79f8767485cf6fb43c7756a8e130a913 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Authenticate user identity via email/SMS verification code. Required before accessing protected resources.",
"parameters_json": "{\"properties\":{\"user_id\":{\"description\":\"Unique client identifier\",\"title\":\"User... | {
"candidate_id": "tool::manage_subscription",
"action_name": "manage_subscription",
"arguments_json": "{\"action\":\"disable_auto_renewal\",\"confirmation\":true,\"user_id\":\"SEA001\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.25,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 403,
"source_identity": "227",
"trajectory_id": "227",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
8dc7f0e46aa42ea4d1571d4ce62425a2971d3ea2db774d56faa7eb5311dc172e | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 5cf7a731425200e1a6ff9550d1b2f0c28e96c722c1cfc0ba68a8b4b88c69859c | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Authenticate user using full name with either email or phone plus security code. Required for profile-level actions.",
"parameters_json": "{\"properties\":{\"email\":{\"description\":\"User's registered email address\",\... | {
"candidate_id": "tool::get_wedding_details",
"action_name": "get_wedding_details",
"arguments_json": "{\"event_id\":\"GC-4320\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.1875,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 404,
"source_identity": "1072",
"trajectory_id": "1072",
"step_index": 3,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
c2ef4b2f045e6a24eea3a3028b05cdb939dd8a3379bc208258fd2435202efdb2 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 0c20b232a41c060177e56c77a34c78c73abd0334451b4316f71a4ea914e0a044 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Verify user identity using:\n- Account ID + registered email\nOR\n- Order number + last 4 digits of payment method",
"parameters_json": "{\"properties\":{\"account_id\":{\"description\":\"User account ID (if using accoun... | {
"candidate_id": "tool::initiate_return",
"action_name": "initiate_return",
"arguments_json": "{\"issue_type\":\"certification_failure\",\"lot_numbers\":[\"LOT-2023-077889\"],\"order_id\":\"O77889\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.46875,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 406,
"source_identity": "25",
"trajectory_id": "25",
"step_index": 1,
"decision_ordinal":... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
9c59456d501512d5d00ed440c79393106d85faf9efb27ae98c49cfe598cbb5d7 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | e4c2b28d6bfb5f9cd05ee96476a927cc0f86cd362611dcc65a0f5f6a2ff876f3 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Authenticate user using full name with either email or phone plus security code. Required for profile-level actions.",
"parameters_json": "{\"properties\":{\"email\":{\"description\":\"User's registered email address\",\... | {
"candidate_id": "tool::process_payment",
"action_name": "process_payment",
"arguments_json": "{\"amount\":5000,\"event_id\":\"GC-4320\",\"payment_phase\":\"booking_deposit\",\"recipient_id\":\"VENUE-2023\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 407,
"source_identity": "1072",
"trajectory_id": "1072",
"step_index": 4,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
f257000a432cb1e8c596871e88c974c449e6ad58d0aec9ba30dd8d7bfc9fb38c | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 0797229abf1b5f92bf1976d897051fc96805939c9f9beac83b2a20d04b76a32d | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_patient",
"name": "authenticate_patient",
"description": "Verify patient identity using date of birth and last four SSN digits or name/phone/ZIP for non-registered users.",
"parameters_json": "{\"properties\":{\"auth_method\":{\"description\":\"Authentication method\",\"e... | {
"candidate_id": "tool::transfer_to_human_agents",
"action_name": "transfer_to_human_agents",
"arguments_json": "{\"summary\":\"Patient P65257 requesting early refill of Suboxone 8mg/2mg (controlled substance). Patient states lost medication while traveling, running out, already used last dose, third refill this... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.1875,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 408,
"source_identity": "639",
"trajectory_id": "639",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
7f2fa26a7856d30c3c4ec48cfdddf1bc789bee7dae66c50ec47d7a9abd5b8c0a | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 6be1b7ba037fd638e5d65307775a8a571f0b9f29d82e6a470dfa041cf47293ad | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_customer",
"name": "authenticate_customer",
"description": "Verify customer identity using Customer ID and service address",
"parameters_json": "{\"properties\":{\"customer_id\":{\"type\":\"string\"},\"service_address\":{\"type\":\"string\"}},\"required\":[\"customer_id\"... | {
"candidate_id": "tool::get_fee_structure",
"action_name": "get_fee_structure",
"arguments_json": "{\"customer_tier\":\"Commercial\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.59375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 409,
"source_identity": "640",
"trajectory_id": "640",
"step_index": 2,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
1dbfbee48a38cf5f2cefb33aabf73eae48455b97025b92e1c196808f783b2fa8 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 2b0f36c9b706f5db3e6419b00d71013a5556f2e3b0c9ecc63b9792679caff8b1 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::search_events",
"name": "search_events",
"description": "Search available events based on date, venue, or artist. Filters events occurring after the current time.",
"parameters_json": "{\"properties\":{\"artist\":{\"description\":\"Performer or production name\",\"title\":\"Artist\",\... | {
"candidate_id": "tool::get_ticket_details",
"action_name": "get_ticket_details",
"arguments_json": "{\"ticket_id\":\"DANCE2025-08-2B-15\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.5,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 410,
"source_identity": "228",
"trajectory_id": "228",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
42d6421343c4beff4484aeab9f1e134697a8b02256c10e10f2628b604d1703d4 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 1b47ce4963ee5c3283172d2e3422bed94c5592a81ead46591e0c554bb67695f2 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::verify_customer_id_email",
"name": "verify_customer_id_email",
"description": "Verify identity using customer ID and registered email.",
"parameters_json": "{\"properties\":{\"customer_id\":{\"type\":\"string\"},\"email\":{\"type\":\"string\"}},\"required\":[\"customer_id\",\"email\"]... | {
"candidate_id": "tool::get_loyalty_details",
"action_name": "get_loyalty_details",
"arguments_json": "{\"customer_id\":\"34587291\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.28125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 411,
"source_identity": "26",
"trajectory_id": "26",
"step_index": 0,
"decision_ordinal":... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
7cf9490a382f1bc5befe2a30acc65cbafa4a1375030e7409a8ecfce140d913e8 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 9ce6d80071496aaab1c6e5f3f45125c23b7eb3f17db0688e9da57c72999e37cd | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::verify_customer_id_email",
"name": "verify_customer_id_email",
"description": "Verify identity using customer ID and registered email.",
"parameters_json": "{\"properties\":{\"customer_id\":{\"type\":\"string\"},\"email\":{\"type\":\"string\"}},\"required\":[\"customer_id\",\"email\"]... | {
"candidate_id": "tool::get_customer_details",
"action_name": "get_customer_details",
"arguments_json": "{\"customer_id\":\"34587291\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.28125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 413,
"source_identity": "26",
"trajectory_id": "26",
"step_index": 1,
"decision_ordinal":... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
a080bec0b8203d54996e310c18b1f003389f1b9b7a67d901e3322fb542432d88 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | a3d2e4ed3db379848c1642625fcae2dbf44ff1292688e65593ba6a200a395f20 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_influencer",
"name": "authenticate_influencer",
"description": "Verify influencer identity using ID and date of birth. Required for accessing influencer-specific functions.",
"parameters_json": "{\"properties\":{\"date_of_birth\":{\"type\":\"string\"},\"influencer_id\":{\... | {
"candidate_id": "tool::modify_deliverables",
"action_name": "modify_deliverables",
"arguments_json": "{\"campaign_id\":\"CAM-3337\",\"new_deliverables\":[{\"quantity\":2,\"type\":\"story\"},{\"quantity\":3,\"type\":\"reel\"},{\"quantity\":2,\"type\":\"post\"}]}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.09375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 414,
"source_identity": "1074",
"trajectory_id": "1074",
"step_index": 0,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
e6cc02b7bc2d7ff9ef42f3b3fa475e9c9858860afd0afc3df23c02786bb12b97 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 06a5bc65390948a084c36701028b84a2eec9ef288c69f50548605000618de3e2 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_influencer",
"name": "authenticate_influencer",
"description": "Verify influencer identity using ID and date of birth. Required for accessing influencer-specific functions.",
"parameters_json": "{\"properties\":{\"date_of_birth\":{\"type\":\"string\"},\"influencer_id\":{\... | {
"candidate_id": "tool::get_campaign_status",
"action_name": "get_campaign_status",
"arguments_json": "{\"campaign_id\":\"CAM-3337\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 416,
"source_identity": "1074",
"trajectory_id": "1074",
"step_index": 1,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
a5ba2c4c6ed654f334d3390e8128afcc59d87984c8601e9b92c9f018f68c556a | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 4626b74dc68c87144f4f79a5aaadd31779c73e02cb28b87640e4daa779726f26 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_client",
"name": "authenticate_client",
"description": "Authenticate client using Client ID and Order ID for order actions or email/phone for profile access",
"parameters_json": "{\"properties\":{\"auth_method\":{\"description\":\"Authentication method to use\",\"enum\":[... | {
"candidate_id": "tool::escalate_to_human",
"action_name": "escalate_to_human",
"arguments_json": "{\"summary\":\"Client HR456 order RELIGIOUS12 requests modification to make entire menu kosher-certified for religious observance. Event in approximately 116 hours. Kosher certification not available as standard me... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.28125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 417,
"source_identity": "427",
"trajectory_id": "427",
"step_index": 3,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
e15eb1ec331a64e69b79822c7fae8adb8746fcc9f3198cdb0d06607212bb40e0 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | a2c2485c56fa748624b4199c43ed782bf586aa6b92e55a3cd7e9b51b4eb9c069 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_customer",
"name": "authenticate_customer",
"description": "Authenticate customer using VIN and phone number verification. Required before any account-sensitive actions.",
"parameters_json": "{\"properties\":{\"phone_number\":{\"description\":\"Registered phone number inc... | {
"candidate_id": "tool::transfer_to_human_agent",
"action_name": "transfer_to_human_agent",
"arguments_json": "{\"summary\":\"Customer Margaret Thompson (CUST-123456) requesting payment plan options for $1,200 repair estimate. Customer is on fixed income and cannot pay full amount upfront. Policy requires full p... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.0625,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 422,
"source_identity": "230",
"trajectory_id": "230",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
49dd182330193cf71b668b536bce145bdacb4686678f22101541e6fa491e45f7 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | af414ad62fdfcdccbfd1c1a89773e155ae2f1495e817f52592c3002e9efcc56a | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Authenticate user using driver's license ID and reservation confirmation code. Required before accessing reservation details or performing actions.",
"parameters_json": "{\"properties\":{\"driver_license_id\":{\"descript... | {
"candidate_id": "tool::transfer_to_human_agents",
"action_name": "transfer_to_human_agents",
"arguments_json": "{\"summary\":\"Customer reports contaminated fuel from Station X on July 19 causing engine damage and $280 repair. Disputes $150 refuel charge. Requests investigation into fuel supply chain and refund... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.46875,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 423,
"source_identity": "642",
"trajectory_id": "642",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
aa817243d765ce486268ee4644c3d64e576660903e4e9a158707f75010aae096 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 519c1793d68c12531fb07c9eab873dd635b8c53be944b9e5f7fcceecf768ecec | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::verify_identity",
"name": "verify_identity",
"description": "Authenticate customer using reservation ID + contact info or driver's license + ZIP code",
"parameters_json": "{\"anyOf\":[{\"required\":[\"reservation_id\",\"contact_info\"]},{\"required\":[\"driver_license\",\"zip_code\"]}... | {
"candidate_id": "tool::verify_identity",
"action_name": "verify_identity",
"arguments_json": "{\"contact_info\":\"your_email@example.com\",\"reservation_id\":\"RES890123\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.5625,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 425,
"source_identity": "429",
"trajectory_id": "429",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
91510d787eef0d6aef789fdc496fa44f284bf25af15b398cce5ff8dbeda36206 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 6942b5b37e5f21d2288b4d2baf2923c4a36faba19472a9cad46cc91e727a0c89 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::create_event_draft",
"name": "create_event_draft",
"description": "Create a new event draft with basic details.",
"parameters_json": "{\"properties\":{\"client_id\":{\"description\":\"Unique client identifier\",\"type\":\"string\"},\"estimated_attendees\":{\"description\":\"Initial he... | {
"candidate_id": "tool::transfer_to_human_agent",
"action_name": "transfer_to_human_agent",
"arguments_json": "{\"summary\":\"Client needs assistance with in-kind donation for catering services under Event ID DONATE25 and Client ID INKIND25. Requires: 1) Verification of signed donation agreement, 2) Blocking tax... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 427,
"source_identity": "232",
"trajectory_id": "232",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
b84e56930d846d86187f9f2d2cc30cd3732f3baefb78e65cb7f30c83cd4abc03 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | a59cca23f78c97fc7e350ebb7b759bf06b927fa22c4f6455b878d108698cb3f4 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::create_customer_profile",
"name": "create_customer_profile",
"description": "Create a new customer profile with contact information and payment methods.",
"parameters_json": "{\"properties\":{\"contact_info\":{\"description\":\"Customer contact details (phone/email)\",\"type\":\"objec... | {
"candidate_id": "tool::get_booking_status",
"action_name": "get_booking_status",
"arguments_json": "{\"booking_id\":\"ESPORTS5\",\"confirmation_code\":\"#ESPORTS5\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 428,
"source_identity": "872",
"trajectory_id": "872",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
2ed433851a0299bfd75db9aa6b3bdda3b5557737a8f66f48b57f1a18f6a0aba7 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | ec39f83758b232b3d307acfa41aaadecd9242bde5d337e86fc82f013d6343e08 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_via_email_and_device",
"name": "authenticate_via_email_and_device",
"description": "Verify user identity using registered email and device serial number",
"parameters_json": "{\"properties\":{\"device_serial\":{\"type\":\"string\"},\"email\":{\"type\":\"string\"}},\"requi... | {
"candidate_id": "tool::verify_device_ownership",
"action_name": "verify_device_ownership",
"arguments_json": "{\"device_serial\":\"A1B2C3D4E5\",\"user_id\":\"user_12345\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.40625,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 430,
"source_identity": "1077",
"trajectory_id": "1077",
"step_index": 2,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
f1abd10941cd179a888117f2bd4f520e7c61338fe5682d754eb0414f8f998b4e | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 9dc5d0bcad53e3b12169c12b013ec5508fe8879b2ee73e159ecc3017eb748743 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::get_farm_details",
"name": "get_farm_details",
"description": "Retrieve farm information including field boundaries, soil profile, and crop cycle.",
"parameters_json": "{\"properties\":{\"farm_id\":{\"description\":\"Unique farm identifier\",\"type\":\"string\"}},\"required\":[\"farm_... | {
"candidate_id": "tool::verify_field_conditions",
"action_name": "verify_field_conditions",
"arguments_json": "{\"field_id\":\"ND-WHEAT-ORG\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.03125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 433,
"source_identity": "873",
"trajectory_id": "873",
"step_index": 1,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
f8087ffddfa51bcc8d06fc79397ffeb63a3724e5d6694e7f0a5fcbff21b52e21 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 05ceba41f098c2313fefbb26688029f0960172e57ff2db82fed253a3ab0214fd | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Authenticate user identity using User ID and email verification. Required for account operations.",
"parameters_json": "{\"properties\":{\"user_id\":{\"description\":\"Unique user identifier\",\"type\":\"string\"},\"veri... | {
"candidate_id": "tool::get_post_status",
"action_name": "get_post_status",
"arguments_json": "{\"post_id\":\"CAREERFAIR23\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.53125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 434,
"source_identity": "30",
"trajectory_id": "30",
"step_index": 0,
"decision_ordinal":... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
d5baed9aa291f400c3e36d1a183ecd405b0d680a2e6cd4b4b7ae2ffc9f764433 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 48cdb712a54b4925c0e59e54b9dffba04bbbbde44be51cdf319c3e2085c90706 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_rider",
"name": "authenticate_rider",
"description": "Authenticate a rider using Rider ID and phone/email verification code.",
"parameters_json": "{\"properties\":{\"rider_id\":{\"description\":\"Unique rider identifier\",\"type\":\"string\"},\"verification_code\":{\"desc... | {
"candidate_id": "tool::book_service",
"action_name": "book_service",
"arguments_json": "{\"duration\":60,\"horse_id\":\"H101\",\"rider_id\":\"R22334\",\"service_type\":\"Lesson\",\"start_time\":\"2025-11-10T10:00:00-05:00\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 436,
"source_identity": "234",
"trajectory_id": "234",
"step_index": 4,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
581c9b7286e104a447f63ac95bc5d4272d6fef8f05abc27e44896eaebcc82baf | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 704f69a2f1680bd292b83ad1d3bfcc15c8d5a92186d7b771600ca882b2a0608e | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::get_farm_details",
"name": "get_farm_details",
"description": "Retrieve farm information including field boundaries, soil profile, and crop cycle.",
"parameters_json": "{\"properties\":{\"farm_id\":{\"description\":\"Unique farm identifier\",\"type\":\"string\"}},\"required\":[\"farm_... | {
"candidate_id": "tool::get_subscription_tier",
"action_name": "get_subscription_tier",
"arguments_json": "{\"account_id\":\"ND-WHEAT-ORG\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.03125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 437,
"source_identity": "873",
"trajectory_id": "873",
"step_index": 2,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
08ee9abd013d80f7d88df7aa425523efd860f69250476315fb88bdf8d6360f9f | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 59f43f556965eb49b2310b0d63546fb98c80c56b8fbc91a76fb4030974a9f2a2 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Authenticate user identity using User ID and email verification. Required for account operations.",
"parameters_json": "{\"properties\":{\"user_id\":{\"description\":\"Unique user identifier\",\"type\":\"string\"},\"veri... | {
"candidate_id": "tool::transfer_to_human_agent",
"action_name": "transfer_to_human_agent",
"arguments_json": "{\"summary\":\"University career services director needs urgent assistance with post ID CAREERFAIR23 stuck in 3-approver bottleneck for October 15th job fair event. User requests discussion with team ad... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.46875,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 438,
"source_identity": "30",
"trajectory_id": "30",
"step_index": 1,
"decision_ordinal":... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
59c7fef22e4706f460715ef54d95ee999448b0cc321a2e04ce416e91f1fed863 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 276b03b6660ecddc00b62799119d86f490581fbe0e338e602451618a0660d49c | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_customer",
"name": "authenticate_customer",
"description": "Authenticate customer using Customer ID and Equipment PIN.\nRequired before any account-sensitive operations.",
"parameters_json": "{\"properties\":{\"customer_id\":{\"description\":\"Unique customer identifier\"... | {
"candidate_id": "tool::transfer_to_human_agents",
"action_name": "transfer_to_human_agents",
"arguments_json": "{\"summary\":\"Customer GT-9921 reports chemical contamination issue with combine serial COM-GT2021-991. Unknown chemical composition. Immediate specialist handling required per safety protocols.\"}",... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.46875,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 439,
"source_identity": "645",
"trajectory_id": "645",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
c56355f504f472fd5b42ca0ad591176e2990822c3c726529f271a10d90a7e781 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | a3db6fc516c5fe101f2e3667969ba58241ebb5a75c2e3e7515c063feac2a93ef | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_customer",
"name": "authenticate_customer",
"description": "Authenticate customer using Account Number + ZIP or Phone + Last 4 SSN. Required before accessing any account data.",
"parameters_json": "{\"properties\":{\"credentials\":{\"anyOf\":[{\"properties\":{\"account_nu... | {
"candidate_id": "tool::request_service_change",
"action_name": "request_service_change",
"arguments_json": "{\"account_number\":\"789123456\",\"action_type\":\"start\",\"effective_date\":\"2025-05-10\",\"military_deployment\":false}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.46875,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 440,
"source_identity": "430",
"trajectory_id": "430",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
0efc07c7967c309914561ac236648ece601de8df5446eafe141b1f05bd402768 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 374aa2585ab32ff279d2e2cfcf53def86388815a9a3935ac282163080eb7d743 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_client",
"name": "authenticate_client",
"description": "Verify client identity using ID and birthdate",
"parameters_json": "{\"properties\":{\"birthdate\":{\"format\":\"date\",\"type\":\"string\"},\"client_id\":{\"type\":\"string\"}},\"required\":[\"client_id\",\"birthdat... | {
"candidate_id": "tool::verify_family_member",
"action_name": "verify_family_member",
"arguments_json": "{\"client_id\":\"C11227\",\"family_member_id\":\"FAM-12345\",\"pin\":\"5678\",\"relationship\":\"Son\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.3125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 442,
"source_identity": "33",
"trajectory_id": "33",
"step_index": 0,
"decision_ordinal":... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
85554e402162cce3be632afa37184ce17f13a7bc2e4b38b10899dff58a3a2ab5 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 4ca6dc2857d6c1d16bada524e07af6a1f268c4d47419a3e611454630a0e19105 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::verify_identity",
"name": "verify_identity",
"description": "Authenticate user identity using allowed methods for sensitive operations",
"parameters_json": "{\"properties\":{\"code\":{\"description\":\"Verification code/PIN\",\"type\":\"string\"},\"method\":{\"description\":\"Authenti... | {
"candidate_id": "tool::check_device_status",
"action_name": "check_device_status",
"arguments_json": "{\"device_id\":\"HUB2023\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.59375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 443,
"source_identity": "431",
"trajectory_id": "431",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
3a305b05da0a17e95a25937b7a7a5bf68550bcc7f3572ac4d1a03af077424bb0 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 7743ca6ccd4bb9d303882b0edb61618729ef2f9ea6394ee03fa5d890573e86f1 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_client",
"name": "authenticate_client",
"description": "Verify client identity using ID and birthdate",
"parameters_json": "{\"properties\":{\"birthdate\":{\"format\":\"date\",\"type\":\"string\"},\"client_id\":{\"type\":\"string\"}},\"required\":[\"client_id\",\"birthdat... | {
"candidate_id": "tool::get_scheduled_services",
"action_name": "get_scheduled_services",
"arguments_json": "{\"client_id\":\"C11227\",\"end_date\":\"2025-12-25\",\"start_date\":\"2025-12-25\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 445,
"source_identity": "33",
"trajectory_id": "33",
"step_index": 3,
"decision_ordinal":... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
fa62a43f478a81a99fb97acd047622536bb3f01148e05c8c270cc436cd4d3a50 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | a1731e936c681ea661ba85e681dd32a1cd69e1be406c134a69fedfcef065f1df | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::verify_identity",
"name": "verify_identity",
"description": "Authenticate user identity using allowed methods for sensitive operations",
"parameters_json": "{\"properties\":{\"code\":{\"description\":\"Verification code/PIN\",\"type\":\"string\"},\"method\":{\"description\":\"Authenti... | {
"candidate_id": "tool::get_subscription_tier",
"action_name": "get_subscription_tier",
"arguments_json": "{\"user_id\":\"MEDICAL1\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 446,
"source_identity": "647",
"trajectory_id": "647",
"step_index": 1,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
df2786be83e67c0a234586c49ca6dddf942ebd50b06ee8f178b3c4836341e4b8 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 9b314684565aa6f4d606a5fdb838b09c32508305b265841ca854118c175d6c9a | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::verify_identity",
"name": "verify_identity",
"description": "Authenticate user identity using allowed methods for sensitive operations",
"parameters_json": "{\"properties\":{\"code\":{\"description\":\"Verification code/PIN\",\"type\":\"string\"},\"method\":{\"description\":\"Authenti... | {
"candidate_id": "tool::verify_identity",
"action_name": "verify_identity",
"arguments_json": "{\"code\":\"4567\",\"method\":\"pin\",\"user_id\":\"MEDICAL1\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.3125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 448,
"source_identity": "647",
"trajectory_id": "647",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
76100b806a93a69c61bcf63e91ac7db03dec27874fd2f446daa3e666c72420e7 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | b18f2d132984b497b6994a232b0828213c54ddf6855b530c476e5c0973bdca1b | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::verify_client_identity",
"name": "verify_client_identity",
"description": "Authenticate client identity through account ID + registered phone, email + security code, or company ID + contact validation",
"parameters_json": "{\"properties\":{\"auth_method\":{\"description\":\"Authentica... | {
"candidate_id": "tool::verify_client_identity",
"action_name": "verify_client_identity",
"arguments_json": "{\"auth_method\":\"email\",\"credentials\":{\"auth_value\":\"enterprise@e8833.com\",\"client_id\":\"E-8833\"}}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.40625,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 450,
"source_identity": "432",
"trajectory_id": "432",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
3e3c7675667a34fb43cf8a3c62ee1fb3979da315a69a4f0d3a37bc95b1db9bf6 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | f680f94b6828d393d00c0e0e1cb795ed11662561f630b8754d2462e814e9c4b3 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_client",
"name": "authenticate_client",
"description": "Verify client identity using ID and birthdate",
"parameters_json": "{\"properties\":{\"birthdate\":{\"format\":\"date\",\"type\":\"string\"},\"client_id\":{\"type\":\"string\"}},\"required\":[\"client_id\",\"birthdat... | {
"candidate_id": "tool::get_client_profile",
"action_name": "get_client_profile",
"arguments_json": "{\"client_id\":\"C11227\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.1875,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 453,
"source_identity": "33",
"trajectory_id": "33",
"step_index": 2,
"decision_ordinal":... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
8bae6458d51ca3fedb93f996d1598c92617b8fea7833660b63086ca42e8291bc | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 125ed687eaf976db18c35383ae7753ea289dff96ecd347718715c23dade8e2fc | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_client",
"name": "authenticate_client",
"description": "Verify client identity using ID and birthdate",
"parameters_json": "{\"properties\":{\"birthdate\":{\"format\":\"date\",\"type\":\"string\"},\"client_id\":{\"type\":\"string\"}},\"required\":[\"client_id\",\"birthdat... | {
"candidate_id": "tool::schedule_service",
"action_name": "schedule_service",
"arguments_json": "{\"client_id\":\"C11227\",\"consent_obtained\":true,\"duration\":2,\"emergency_surcharge\":true,\"medication_flag\":true,\"service_type\":\"DailyLiving\",\"start_time\":\"2025-12-25 08:00:00\"}",
"label_json": null... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 454,
"source_identity": "33",
"trajectory_id": "33",
"step_index": 0,
"decision_ordinal":... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
af1d30f25e6c47dd0aa8d3d109508a0fd113e2ddb438b5052e194adfa82f76f6 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | d2b9657c9260f766ea7c9869a7a04fa137ca5e385bd5555cdd9eecf5ccdcdf44 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::check_attraction_status",
"name": "check_attraction_status",
"description": "Retrieve current operational status and details of a theme park attraction",
"parameters_json": "{\"properties\":{\"attractionId\":{\"description\":\"Unique identifier for the attraction\",\"type\":\"string\"... | {
"candidate_id": "tool::transfer_to_human_agent",
"action_name": "transfer_to_human_agent",
"arguments_json": "{\"summary\":\"Senior guest SENIOR-9900 with ticket TKT-2025-1100 requests membership freeze due to sudden illness, unable to attend scheduled visit on May 20th. Requires medical documentation review fo... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.5,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 455,
"source_identity": "433",
"trajectory_id": "433",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
d3a434637080ac2861c75ce08dc1626fa673245e88307ee48fec398b5c363b6a | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | ca8697d7406f838a84387a9c5262b6c2c1177239846354e94fcaaaeec99b8954 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Verify user identity using User ID plus registered email or phone",
"parameters_json": "{\"properties\":{\"email\":{\"description\":\"Registered email address\",\"title\":\"Email\",\"type\":\"string\"},\"phone\":{\"descr... | {
"candidate_id": "tool::get_policy_information",
"action_name": "get_policy_information",
"arguments_json": "{\"policy_topic\":\"refund\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.5,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 460,
"source_identity": "239",
"trajectory_id": "239",
"step_index": 1,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
168cf508e8e03c594aef5c82c9566142748c2ed00cdb464c2a86dedb32eb4be2 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 2aedc01a6d18c24bd8b6382872c641762770bc641c312aefec4b7c663cfa9c5f | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Authenticate user identity using:\n- Phone number + license plate\n- Email address + license plate",
"parameters_json": "{\"properties\":{\"email\":{\"description\":\"User's email address\",\"type\":\"string\"},\"license... | {
"candidate_id": "tool::book_appointment",
"action_name": "book_appointment",
"arguments_json": "{\"service_type\":\"Full Detailing\",\"time_slot\":\"2024-01-20T11:00:00Z\",\"user_id\":\"user_12345\",\"vehicle_license_plate\":\"PAINT3\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.03125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 463,
"source_identity": "1081",
"trajectory_id": "1081",
"step_index": 1,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
aa0871113c3443a9a470a11b0c7f28474d6d68a9cb1957ac72f55540523f6363 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | fa743f3dbe1bc911964aab1ecb8ccae442c745a2b556a624deef15ad99eaac86 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Authenticate user using either:\n\n- User ID + Registered Email + Last 4 Phone Digits, OR\n- Order ID + Billing ZIP Code",
"parameters_json": "{\"properties\":{\"auth_method\":{\"description\":\"Authentication method typ... | {
"candidate_id": "tool::authenticate_user",
"action_name": "authenticate_user",
"arguments_json": "{\"auth_method\":\"order_credentials\",\"credentials\":{\"billing_zip_code\":\"94043\",\"order_id\":\"PN-LAWLIB-8901\"}}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.34375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 464,
"source_identity": "36",
"trajectory_id": "36",
"step_index": 0,
"decision_ordinal":... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
70546e0b1272b560fb210d3b36140f52c29143e31965ea1b643f0e217d80e8d2 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 93f291a2e89c2e8c8dc372e99e7f0d5d2475c562183d91148e4a09b0a7e27d36 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::get_user_details",
"name": "get_user_details",
"description": "Retrieve user profile details including delivery addresses, payment methods, loyalty tier, substitution preferences, and dietary restrictions.",
"parameters_json": "{\"properties\":{\"user_id\":{\"description\":\"Unique us... | {
"candidate_id": "tool::transfer_to_human_agent",
"action_name": "transfer_to_human_agent",
"arguments_json": "{\"summary\":\"User has IBS and needs to confirm FODMAP levels of items in order #FX88408 using Monash University certification standards, as standard dietary flags aren't sufficient for their medical n... | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.5625,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 465,
"source_identity": "653",
"trajectory_id": "653",
"step_index": 0,
"decision_ordinal... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
304fff1d7ef41c61194873cd838fa88d95df1de3c54de612006c38c1bf5dde53 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | e00089c70c6273ce14c5597873b08da0ad494e403d4e6d2a22f500a75fd5306f | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::generate_quote",
"name": "generate_quote",
"description": "Generate instant translation quote based on project specifications",
"parameters_json": "{\"properties\":{\"document_type\":{\"description\":\"Type of document being translated\",\"enum\":[\"legal\",\"medical\",\"technical\",\... | {
"candidate_id": "tool::get_project_status",
"action_name": "get_project_status",
"arguments_json": "{\"project_id\":\"ACKNOW77\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.59375,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 466,
"source_identity": "1082",
"trajectory_id": "1082",
"step_index": 2,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
6d99f48ab7b37f68e1878b20fb84f56a5abeabe57d6486bf2c8cfd23c74d33d8 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 53022ad4cbdfd3973ba3dee04aa0215e02a632261e60c8d5514d8092f62c4058 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::generate_quote",
"name": "generate_quote",
"description": "Generate instant translation quote based on project specifications",
"parameters_json": "{\"properties\":{\"document_type\":{\"description\":\"Type of document being translated\",\"enum\":[\"legal\",\"medical\",\"technical\",\... | {
"candidate_id": "tool::get_client_profile",
"action_name": "get_client_profile",
"arguments_json": "{\"client_id\":\"C-77291\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.3125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 468,
"source_identity": "1082",
"trajectory_id": "1082",
"step_index": 3,
"decision_ordin... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
75fb0db31c4d5c987f7ba71a188c021b7f9e86ee72e903cd933825c658d43506 | nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1 | tool_choice | trainable | null | 7731f2baee8d26db9c273dba4fa81f5fee9efa0f5f281ff03e578852304d13f0 | {
"system": "You are a customer service agent that helps the user. The policy that determines how you should respond to requests from users is described below between the <policy> and </policy> tags.\n\nIn each turn you can either:\n- Send a message to the user.\n- Make a tool call.\nYou cannot do both at the same t... | [
{
"id": "tool::authenticate_user",
"name": "authenticate_user",
"description": "Authenticate user using either:\n\n- User ID + Registered Email + Last 4 Phone Digits, OR\n- Order ID + Billing ZIP Code",
"parameters_json": "{\"properties\":{\"auth_method\":{\"description\":\"Authentication method typ... | {
"candidate_id": "tool::get_order_status",
"action_name": "get_order_status",
"arguments_json": "{\"order_id\":\"PN-LAWLIB-8901\"}",
"label_json": null
} | [] | {
"complete": null,
"trajectory_success": null,
"value_target": null,
"reward": null,
"score": null,
"source_pass_rate": 0.28125,
"outcome_evidence": null
} | {
"dataset_id": "nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1",
"dataset_revision": "9643c8103d7bfbc2d7fc4d15991d6739c612ff58",
"source_config": "default",
"raw_split": "train",
"source_row_index": 469,
"source_identity": "36",
"trajectory_id": "36",
"step_index": 1,
"decision_ordinal":... | {
"choice_eligible": true,
"completion_eligible": false,
"value_eligible": false,
"score_eligible": false,
"boolean_eligible": false,
"arguments_eligible": true,
"use_for_bc": true,
"use_for_value": false,
"supervision_evidence": "explicit_expected_action",
"choice_ineligible_reason": null,
"quali... | train |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.