Preference Alignment on MedQA
77.4Preference AlignmentPep
Evaluation results
| Method | Links | |
|---|---|---|
| Pep2026.02 | 77.4 | |
| Pop. Average2026.02 | 73.2 | |
| GRPO2026.02 | 71.4 | |
| Prompting2026.02 | 22.3 | |
| CollabLLM2026.02 | 20.3 |
| Method | Links | |
|---|---|---|
| Pep2026.02 | 77.4 | |
| Pop. Average2026.02 | 73.2 | |
| GRPO2026.02 | 71.4 | |
| Prompting2026.02 | 22.3 | |
| CollabLLM2026.02 | 20.3 |