Medical Question Answering on Medical QA offline evaluation
0.83HonestyUni-Normalize Scheme
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Uni-Normalize SchemeAlignment Scheme=Uni-Normalize2026.02 | 0.83 | 0.935 | 0.225 | 573.1 | |
| Uni-Normalize Scheme + Like/Dislike Helpful RMAlignment Scheme=Uni-Normalize + Like/Dislike Helpful RM2026.02 | 0.779 | 0.94 | 0.206 | 412 | |
| BaselineAlignment Scheme=Baseline2026.02 | 0.77 | 0.94 | 0.165 | 424.6 |