Medical Domain Question Answering on RaR-Medicine In-domain Unseen
20.22ScorePrompt
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| PromptModel=Deepseek-V4-Flash, Category=Vanilla2026.06 | 20.22 | 75 | |
| PromptModel=Qwen3-32B, Category=Vanilla2026.06 | 20.05 | 74.69 | |
| EvoRubricsModel=Qwen3-8B, Category=Ours2026.06 | 17.34 | 61.71 | |
| PromptModel=Qwen3-8B, Category=Vanilla2026.06 | 17 | 60.51 | |
| EvoRubricsModel=Qwen3-4B, Category=Ours2026.06 | 16.36 | 60.5 | |
| OnlineRubricsModel=Qwen3-8B, Category=RL w/ Rubric2026.06 | 16.32 | 57.85 | |
| GoldenRubricsModel=Qwen3-8B, Category=RL w/ Rubric2026.06 | 16.16 | 57.26 | |
| SFTModel=Qwen3-8B, Category=Supervised2026.06 | 15.94 | 56.5 | |
| PromptModel=Qwen3-4B, Category=Vanilla2026.06 | 15.14 | 56.04 | |
| RuscaRLModel=Qwen3-8B, Category=RL w/ Rubric2026.06 | 12.67 | 44.82 | |
| SFTModel=Qwen3-4B, Category=Supervised2026.06 | 12.35 | 45.67 | |
| GoldenRubricsModel=Qwen3-4B, Category=RL w/ Rubric2026.06 | 11.41 | 42.16 | |
| RuscaRLModel=Qwen3-4B, Category=RL w/ Rubric2026.06 | 10.98 | 40.94 | |
| OnlineRubricsModel=Qwen3-4B, Category=RL w/ Rubric2026.06 | 10.48 | 38.66 |