Medical Question Answering on MedXpertQA-R
11.12Pass@1SHIFT
Evaluation Results
| Method | Links | |
|---|---|---|
| SHIFTTrain Size=0.2%2026.05 | 11.12 | |
| Qwen3-1.7BTrain Size=100%2026.05 | 10.26 | |
| SHIFTTrain Size=0.1%2026.05 | 9.62 | |
| ClusterTrain Size=0.2%2026.05 | 9.35 | |
| CoT SimilarityTrain Size=0.1%2026.05 | 9.19 | |
| CoT SimilarityTrain Size=0.2%2026.05 | 9.03 | |
| Random (avg)Train Size=0.2%2026.05 | 7.88 | |
| SC-EntropyTrain Size=0.1%2026.05 | 7.79 | |
| A-PPLTrain Size=0.2%2026.05 | 7.47 | |
| SC-EntropyTrain Size=0.2%2026.05 | 7.25 | |
| CoreSetTrain Size=0.2%2026.05 | 7.15 | |
| ClusterTrain Size=0.1%2026.05 | 6.93 | |
| Random (avg)Train Size=0.1%2026.05 | 6.84 | |
| Q-PPLTrain Size=0.2%2026.05 | 6.29 | |
| CoreSetTrain Size=0.1%2026.05 | 6.02 | |
| Q-PPLTrain Size=0.1%2026.05 | 5.8 | |
| A-PPLTrain Size=0.1%2026.05 | 4.51 | |
| Qwen3-1.7BTrain Size=N/A2026.05 | 2.36 |