Medical Question Answering on MedQA-USMLE (Accuracy, F1, Precision, Recall)
56.95AccuracyQwen-2.5-7B
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Qwen-2.5-7BData Format=plaintext2026.06 | 56.95 | 56.84 | 56.91 | 56.95 | |
| Llama-3.1-Aloe-Beta-8BData Format=plaintext2026.06 | 56.25 | 56.12 | 56.6 | 56.25 | |
| Llama-3.1-8BData Format=plaintext2026.06 | 53.26 | 53.3 | 53.49 | 53.26 | |
| Llama-3-Med42-8BData Format=plaintext2026.06 | 51.99 | 50.2 | 54.23 | 51.99 | |
| Mistral-7B-Instruct-v0.3Data Format=plaintext2026.06 | 47.09 | 47.07 | 47.17 | 47.09 |