Medical Reasoning on MedBullets 5-option multiple choice
53.9AccuracyHuatuo-o1
Evaluation Results
| Method | Links | |
|---|---|---|
| Huatuo-o1Backbone=LLaMA-3.1-8B-Instruct2026.02 | 53.9 | |
| MedVerseBackbone=LLaMA-3.1-8B-Instruct2026.02 | 53.6 | |
| MedReasonBackbone=LLaMA-3.1-8B-Instruct2026.02 | 51 | |
| MedVerseBackbone=Qwen2.5-7B-Instruct2026.02 | 48 | |
| MedReasonBackbone=Qwen2.5-7B-Instruct2026.02 | 44.2 | |
| Original (LLaMA-3.1-8B-Instruct)Backbone=LLaMA-3.1-8B-Instruct2026.02 | 42.5 | |
| Original (Qwen2.5-7B-Instruct)Backbone=Qwen2.5-7B-Instruct2026.02 | 39.6 |