Medical Reasoning on Lancet
63.3AccuracyEmbedding Diversity
Evaluation results
| Method | Links | |
|---|---|---|
| Embedding DiversityModel=Qwen2.5-7B-Instruct, Training=Finetuning, Sample size=1000 examples2026.06 | 63.3 | |
| S2LModel=Qwen2.5-7B-Instruct, Training=Finetuning, Sample size=1000 examples2026.06 | 62.6 | |
| Middle PerplexityModel=Qwen2.5-7B-Instruct, Training=Finetuning, Sample size=1000 examples2026.06 | 61.4 | |
| OursModel=Qwen2.5-7B-Instruct, Training=Finetuning, Sample size=1000 examples2026.06 | 61.4 | |
| RandomModel=Qwen2.5-7B-Instruct, Training=Finetuning, Sample size=1000 examples2026.06 | 61.2 | |
| m1kModel=Qwen2.5-7B-Instruct, Training=Finetuning, Sample size=1000 examples2026.06 | 60.4 | |
| LearnabilityModel=Qwen2.5-7B-Instruct, Training=Finetuning, Sample size=1000 examples2026.06 | 59.8 |