Medical Question Answering on MedQA OOD
57.5MCQ AccuracyProMed(S#1+2)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| ProMed(S#1+2)Training=CMB-trained, Backbone=LLaMA-8B, Method Variant=S#1+22025.08 | 57.5 | 25.37 | |
| ProMed(S#1)+GRPOTraining=CMB-trained, Backbone=LLaMA-8B, Method Variant=S#1 + GRPO2025.08 | 51.29 | 21.92 | |
| ProMed(S#1)+DPOTraining=CMB-trained, Backbone=LLaMA-8B, Method Variant=S#1 + DPO2025.08 | 49.88 | 21.29 | |
| VanillaTraining=CMB-trained, Backbone=LLaMA-8B2025.08 | 40.11 | 20.35 |