List your company

Regulatory Reasoning and Claim Validation on AIReg-Bench N=120 systems

75.9AccuracyOur Student-Teacher Methods
65.568.270.973.6Feb 12, 2026

Evaluation results

MethodLinks
Our Student-Teacher MethodsAveraging method=Standard/Micro-Avg2026.0275.997.572.18343.8
Gemini 2.5 ProAveraging method=Standard/Micro-Avg2026.0265.988.765.175.134.4