Medical Reasoning on NEJM
60.8AccuracyOurs
Evaluation results
| Method | Links | |
|---|---|---|
| OursModel=Qwen2.5-7B-Instruct, Training=Finetuning, Sample size=1000 examples2026.06 | 60.8 | |
| m1kModel=Qwen2.5-7B-Instruct, Training=Finetuning, Sample size=1000 examples2026.06 | 60.8 | |
| S2LModel=Qwen2.5-7B-Instruct, Training=Finetuning, Sample size=1000 examples2026.06 | 59.5 | |
| RandomModel=Qwen2.5-7B-Instruct, Training=Finetuning, Sample size=1000 examples2026.06 | 58.9 | |
| Embedding DiversityModel=Qwen2.5-7B-Instruct, Training=Finetuning, Sample size=1000 examples2026.06 | 58.5 | |
| LearnabilityModel=Qwen2.5-7B-Instruct, Training=Finetuning, Sample size=1000 examples2026.06 | 57.8 | |
| Middle PerplexityModel=Qwen2.5-7B-Instruct, Training=Finetuning, Sample size=1000 examples2026.06 | 57.8 |