Medical Reasoning on MMLU OOD
86.6AccuracyPRA (Ours)
Evaluation Results
| Method | Links | |
|---|---|---|
| PRA (Ours)Policy=Qwen3-4B-Instruct2026.04 | 86.6 | |
| RAG + SCPolicy=Qwen3-4B-Instruct2026.04 | 86.2 | |
| RAGPolicy=Qwen3-4B-Instruct2026.04 | 85.4 | |
| CoT + SCPolicy=Qwen3-4B-Instruct2026.04 | 84.9 | |
| CoTPolicy=Qwen3-4B-Instruct2026.04 | 83.7 | |
| DirectPolicy=Qwen3-4B-Instruct2026.04 | 77.4 | |
| Direct + SCPolicy=Qwen3-4B-Instruct2026.04 | 77.3 |