Medical Knowledge Evaluation on DiagnosisArena
61.11AccuracyDOS-CPT
Evaluation Results
| Method | Links | |
|---|---|---|
| DOS-CPTBackbone=Qwen3-14B-Base, CPT Strategy=Distance-to-Optimum Selection2025.12 | 61.11 | |
| RS-CPTBackbone=Qwen3-14B-Base, CPT Strategy=Random Data Sampling2025.12 | 56.4 | |
| LPS-CPTBackbone=Qwen3-14B-Base, CPT Strategy=Low-PPL Sampling2025.12 | 56.4 | |
| HPS-CPTBackbone=Qwen3-14B-Base, CPT Strategy=High-PPL Sampling2025.12 | 45.7 | |
| Qwen3-14B-BaseBackbone=Qwen3-14B-Base, CPT Strategy=None2025.12 | 41.3 |