{ "model": "Qwen/Qwen3.6-35B-A3B-FP8", "api_base": "http://localhost:8000/v1", "timestamp": "2026-05-04T18:30:28.713574", "total_evaluated": 60, "summary": { "total_episodes": 60, "detection_accuracy": 0.5666666666666667, "miss_rate": 0.32142857142857145, "false_alarm_rate": 0.53125, "precision": 0.5277777777777778, "recall": 0.6785714285714286, "f1_security": 0.5937499999999999, "threat_type_accuracy": 0.47368421052631576, "parse_failure_rate": 0.0, "breakdown": { "TP": { "total": 28, "correct": 19, "accuracy": 0.6785714285714286 }, "FP": { "total": 28, "correct": 13, "accuracy": 0.4642857142857143 }, "TN": { "total": 4, "correct": 2, "accuracy": 0.5 } }, "per_sq": { "SQ1": { "total": 13, "accuracy": 0.8461538461538461, "miss_count": 2, "false_alarm_count": 0 }, "SQ2": { "total": 6, "accuracy": 0.5, "miss_count": 2, "false_alarm_count": 1 }, "SQ3": { "total": 17, "accuracy": 0.47058823529411764, "miss_count": 2, "false_alarm_count": 7 }, "SQ4": { "total": 17, "accuracy": 0.35294117647058826, "miss_count": 3, "false_alarm_count": 8 }, "SQ5": { "total": 7, "accuracy": 0.8571428571428571, "miss_count": 0, "false_alarm_count": 1 } } }, "errors": { "total_errors": 36, "error_distribution": { "MISS": 9, "FALSE_ALARM": 17, "WRONG_TYPE": 10 }, "error_by_sq": { "SQ1": { "MISS": 2 }, "SQ2": { "FALSE_ALARM": 1, "MISS": 2 }, "SQ3": { "FALSE_ALARM": 7, "MISS": 2, "WRONG_TYPE": 5 }, "SQ4": { "FALSE_ALARM": 8, "WRONG_TYPE": 4, "MISS": 3 }, "SQ5": { "FALSE_ALARM": 1, "WRONG_TYPE": 1 } }, "error_by_category": { "device_fault": 2, "intrusion": 4, "water_damage": 1, "behavioral_anomaly": 4, "child_specific": 4, "elderly_specific": 2, "fire_gas": 2 } }, "pipeline": "EGPv2" }