{ "model": "Qwen/Qwen3.6-35B-A3B-FP8", "api_base": "http://localhost:8000/v1", "timestamp": "2026-05-05T03:36:19.058898", "total_evaluated": 60, "summary": { "total_episodes": 60, "detection_accuracy": 0.6333333333333333, "miss_rate": 0.6428571428571429, "false_alarm_rate": 0.125, "precision": 0.7142857142857143, "recall": 0.35714285714285715, "f1_security": 0.4761904761904762, "threat_type_accuracy": 0.4, "parse_failure_rate": 0.0, "breakdown": { "TP": { "total": 28, "correct": 10, "accuracy": 0.35714285714285715 }, "FP": { "total": 28, "correct": 24, "accuracy": 0.8571428571428571 }, "TN": { "total": 4, "correct": 4, "accuracy": 1.0 } }, "per_sq": { "SQ1": { "total": 13, "accuracy": 0.5384615384615384, "miss_count": 6, "false_alarm_count": 0 }, "SQ2": { "total": 6, "accuracy": 0.5, "miss_count": 2, "false_alarm_count": 1 }, "SQ3": { "total": 17, "accuracy": 0.5294117647058824, "miss_count": 5, "false_alarm_count": 3 }, "SQ4": { "total": 17, "accuracy": 0.7058823529411765, "miss_count": 5, "false_alarm_count": 0 }, "SQ5": { "total": 7, "accuracy": 1.0, "miss_count": 0, "false_alarm_count": 0 } } }, "errors": { "total_errors": 28, "error_distribution": { "MISS": 18, "FALSE_ALARM": 4, "WRONG_TYPE": 6 }, "error_by_sq": { "SQ1": { "MISS": 6 }, "SQ2": { "FALSE_ALARM": 1, "MISS": 2 }, "SQ3": { "FALSE_ALARM": 3, "MISS": 5, "WRONG_TYPE": 3 }, "SQ4": { "MISS": 5, "WRONG_TYPE": 1 }, "SQ5": { "WRONG_TYPE": 2 } }, "error_by_category": { "device_fault": 6, "intrusion": 4, "water_damage": 1, "behavioral_anomaly": 4, "child_specific": 4, "elderly_specific": 3, "fire_gas": 2 } }, "pipeline": "EGPv2.1" }