List your company

Medical Reasoning on MMLU OOD

86.6AccuracyPRA (Ours)
76.92879.43981.9584.461Apr 10, 2026

Evaluation results

MethodLinks
PRA (Ours)Policy=Qwen3-4B-Instruct2026.0486.6
RAG + SCPolicy=Qwen3-4B-Instruct2026.0486.2
RAGPolicy=Qwen3-4B-Instruct2026.0485.4
CoT + SCPolicy=Qwen3-4B-Instruct2026.0484.9
CoTPolicy=Qwen3-4B-Instruct2026.0483.7
DirectPolicy=Qwen3-4B-Instruct2026.0477.4
Direct + SCPolicy=Qwen3-4B-Instruct2026.0477.3