{ "model": "Qwen/Qwen3.6-35B-A3B-FP8", "api_base": "http://localhost:8000/v1", "timestamp": "2026-05-05T01:36:32.006189", "total_evaluated": 60, "summary": { "total_episodes": 60, "detection_accuracy": 0.65, "miss_rate": 0.5714285714285714, "false_alarm_rate": 0.15625, "precision": 0.7058823529411765, "recall": 0.42857142857142855, "f1_security": 0.5333333333333333, "threat_type_accuracy": 0.5833333333333334, "parse_failure_rate": 0.0, "breakdown": { "TP": { "total": 28, "correct": 12, "accuracy": 0.42857142857142855 }, "FP": { "total": 28, "correct": 25, "accuracy": 0.8928571428571429 }, "TN": { "total": 4, "correct": 2, "accuracy": 0.5 } }, "per_sq": { "SQ1": { "total": 13, "accuracy": 0.6153846153846154, "miss_count": 4, "false_alarm_count": 1 }, "SQ2": { "total": 6, "accuracy": 0.6666666666666666, "miss_count": 2, "false_alarm_count": 0 }, "SQ3": { "total": 17, "accuracy": 0.6470588235294118, "miss_count": 5, "false_alarm_count": 1 }, "SQ4": { "total": 17, "accuracy": 0.5294117647058824, "miss_count": 5, "false_alarm_count": 3 }, "SQ5": { "total": 7, "accuracy": 1.0, "miss_count": 0, "false_alarm_count": 0 } } }, "errors": { "total_errors": 26, "error_distribution": { "FALSE_ALARM": 5, "MISS": 16, "WRONG_TYPE": 5 }, "error_by_sq": { "SQ1": { "FALSE_ALARM": 1, "MISS": 4 }, "SQ2": { "MISS": 2 }, "SQ3": { "FALSE_ALARM": 1, "MISS": 5, "WRONG_TYPE": 2 }, "SQ4": { "FALSE_ALARM": 3, "MISS": 5, "WRONG_TYPE": 1 }, "SQ5": { "WRONG_TYPE": 2 } }, "error_by_category": { "device_fault": 4, "intrusion": 4, "water_damage": 1, "behavioral_anomaly": 3, "child_specific": 3, "elderly_specific": 3, "fire_gas": 3 } }, "pipeline": "EGPv2.1" }