CuriousDragon's picture
Upload folder using huggingface_hub
3df3f87 verified
Raw History Blame Contribute Delete
7.53 kB
{
"arm": "rule",
"metrics": {
"action_accuracy": 0.716,
"argument_exact_match": 0.2785,
"counts": {
"clarify_expected": 26,
"predicted_tool_calls": 37,
"text_expected": 178,
"tool_expected": 79
},
"fabricated_response_rate": 0.0,
"gates": {
"argument_exact_match": {
"pass": false,
"target": 0.98,
"value": 0.2785
},
"fabricated_response_rate": {
"pass": true,
"target": 0.01,
"value": 0.0
},
"invalid_tool_call_rate": {
"pass": true,
"target": 0.01,
"value": 0.0
},
"missing_field_exact_match": {
"pass": false,
"target": 0.9,
"value": 0.1154
},
"response_coverage": {
"pass": false,
"target": 1.0,
"value": 0.9719
},
"task_success_rate": {
"pass": false,
"target": 0.95,
"value": 0.677
},
"tool_selection_accuracy": {
"pass": false,
"target": 0.99,
"value": 0.3924
}
},
"invalid_tool_call_rate": 0.0,
"missing_field_exact_match": 0.1154,
"n_records": 257,
"response_coverage": 0.9719,
"task_success_rate": 0.677,
"tool_selection_accuracy": 0.3924
},
"n_records": 257,
"per_family": {
"ABBREVIATIONS": {
"n": 1,
"success": 1.0
},
"ACCOUNT_HELP": {
"n": 2,
"success": 1.0
},
"BILLING_INVOICE": {
"n": 2,
"success": 1.0
},
"CLAIM_DAMAGE": {
"n": 2,
"success": 1.0
},
"COMPANY_ABOUT": {
"n": 2,
"success": 1.0
},
"COMPANY_ALLIANCE": {
"n": 2,
"success": 1.0
},
"COMPANY_AVIATION": {
"n": 2,
"success": 1.0
},
"COMPANY_CAREERS": {
"n": 2,
"success": 1.0
},
"COMPANY_CERTIFICATIONS": {
"n": 2,
"success": 1.0
},
"COMPANY_ECOMMERCE": {
"n": 2,
"success": 1.0
},
"COMPANY_HEADQUARTERS": {
"n": 2,
"success": 1.0
},
"COMPANY_HISTORY": {
"n": 2,
"success": 1.0
},
"COMPANY_INDUSTRIES": {
"n": 2,
"success": 1.0
},
"COMPANY_INTERNATIONAL": {
"n": 2,
"success": 1.0
},
"COMPANY_LEADERSHIP": {
"n": 2,
"success": 1.0
},
"COMPANY_NETWORK": {
"n": 2,
"success": 0.5
},
"COMPANY_PERFORMANCE": {
"n": 2,
"success": 1.0
},
"COMPANY_REAL_OR_DEMO": {
"n": 2,
"success": 1.0
},
"COMPANY_SIZE": {
"n": 2,
"success": 1.0
},
"COMPANY_SUPPLY_CHAIN": {
"n": 2,
"success": 1.0
},
"COMPANY_SUSTAINABILITY": {
"n": 2,
"success": 1.0
},
"COMPANY_TECHNOLOGY": {
"n": 2,
"success": 1.0
},
"COMPANY_WAREHOUSING": {
"n": 2,
"success": 1.0
},
"COMPLAINT_DELIVERY_ISSUE": {
"n": 2,
"success": 1.0
},
"CONFLICTING_INFO": {
"n": 2,
"success": 1.0
},
"CUSTOMS_DUTIES": {
"n": 2,
"success": 1.0
},
"DELIVERY_OPTIONS": {
"n": 1,
"success": 1.0
},
"GOODBYE": {
"n": 2,
"success": 1.0
},
"GREETING": {
"n": 2,
"success": 1.0
},
"GUIDE_CONDITIONS": {
"n": 4,
"success": 0.5
},
"GUIDE_DANGEROUS": {
"n": 4,
"success": 0.5
},
"GUIDE_DOCUMENTS": {
"n": 4,
"success": 0.5
},
"GUIDE_MISSING_TOPIC": {
"n": 2,
"success": 0.0
},
"GUIDE_OVERVIEW": {
"n": 4,
"success": 0.5
},
"GUIDE_PACKING": {
"n": 4,
"success": 1.0
},
"GUIDE_RESTRICTED": {
"n": 4,
"success": 1.0
},
"HELP_CAPABILITIES": {
"n": 2,
"success": 1.0
},
"HINGLISH": {
"n": 2,
"success": 0.5
},
"INSURANCE_COVERAGE": {
"n": 2,
"success": 1.0
},
"ITEM_ELIGIBILITY": {
"n": 2,
"success": 0.5
},
"LOCATION_BY_PIN": {
"n": 4,
"success": 0.75
},
"LOCATION_DIRECT": {
"n": 4,
"success": 0.75
},
"LOCATION_MISSING_PLACE": {
"n": 2,
"success": 0.5
},
"LOCATION_NOT_FOUND": {
"n": 4,
"success": 0.5
},
"LOCATION_SERVICE_FILTER": {
"n": 4,
"success": 0.5
},
"LOCATION_TIMEOUT": {
"n": 4,
"success": 0.5
},
"LOCATION_WRONG_PLACE_RECOVERY": {
"n": 8,
"success": 0.25
},
"MULTITURN_CORRECTION": {
"n": 2,
"success": 0.0
},
"MULTITURN_REFERENCE": {
"n": 2,
"success": 0.0
},
"MULTITURN_SLOT_FILL": {
"n": 6,
"success": 0.333
},
"MULTITURN_TOPIC_CHANGE": {
"n": 2,
"success": 1.0
},
"OUT_OF_SCOPE": {
"n": 2,
"success": 1.0
},
"PACKAGING_ADVICE": {
"n": 4,
"success": 1.0
},
"PICKUP_SCHEDULING": {
"n": 4,
"success": 0.75
},
"PRICING_POLICY": {
"n": 2,
"success": 1.0
},
"PROMPT_INJECTION": {
"n": 2,
"success": 1.0
},
"PROOF_OF_DELIVERY": {
"n": 4,
"success": 0.75
},
"QUOTE_DIRECT": {
"n": 4,
"success": 0.5
},
"QUOTE_INTERNATIONAL": {
"n": 4,
"success": 0.5
},
"QUOTE_INVALID_WEIGHT": {
"n": 2,
"success": 0.0
},
"QUOTE_MISSING_DESTINATION": {
"n": 2,
"success": 0.0
},
"QUOTE_MISSING_MULTIPLE": {
"n": 2,
"success": 1.0
},
"QUOTE_MISSING_ORIGIN": {
"n": 2,
"success": 0.0
},
"QUOTE_MISSING_WEIGHT": {
"n": 2,
"success": 0.0
},
"QUOTE_UNAVAILABLE": {
"n": 4,
"success": 0.5
},
"QUOTE_WRONG_WEIGHT_RECOVERY": {
"n": 6,
"success": 0.333
},
"REDELIVERY_ADDRESS_CHANGE": {
"n": 2,
"success": 1.0
},
"REQUEST_TO_FABRICATE": {
"n": 2,
"success": 1.0
},
"REQUEST_TO_GUESS": {
"n": 2,
"success": 1.0
},
"RETURNS_REVERSE": {
"n": 2,
"success": 1.0
},
"SERVICE_DETAIL": {
"n": 4,
"success": 0.5
},
"SERVICE_LIST": {
"n": 4,
"success": 0.5
},
"SERVICE_POINT_LOCKER": {
"n": 4,
"success": 0.75
},
"SMALLTALK": {
"n": 2,
"success": 1.0
},
"SUPPORT_CONTACT": {
"n": 4,
"success": 0.5
},
"SUPPORT_FAQ": {
"n": 4,
"success": 0.5
},
"SUPPORT_NEWS": {
"n": 4,
"success": 0.5
},
"THANKS": {
"n": 2,
"success": 1.0
},
"TOOL_INJECTION": {
"n": 2,
"success": 0.5
},
"TRACK_DIRECT": {
"n": 4,
"success": 1.0
},
"TRACK_INVALID_FORMAT": {
"n": 2,
"success": 0.0
},
"TRACK_MISSING_NUMBER": {
"n": 2,
"success": 0.0
},
"TRACK_MULTIPLE": {
"n": 4,
"success": 0.5
},
"TRACK_NOT_FOUND": {
"n": 4,
"success": 1.0
},
"TRACK_REFERENCE": {
"n": 4,
"success": 1.0
},
"TRACK_STATUS_CONTRADICTION": {
"n": 4,
"success": 1.0
},
"TRACK_TIMEOUT": {
"n": 4,
"success": 0.5
},
"TRACK_UNAVAILABLE": {
"n": 4,
"success": 0.5
},
"TRACK_WRONG_NUMBER_RECOVERY": {
"n": 8,
"success": 0.375
},
"TYPOS": {
"n": 1,
"success": 1.0
}
}
}