List your company

Medical Knowledge Evaluation on DiagnosisArena

61.11AccuracyDOS-CPT
40.507645.856351.20556.5537Dec 25, 2025

Evaluation results

MethodLinks
DOS-CPTBackbone=Qwen3-14B-Base, CPT Strategy=Distance-to-Optimum Selection2025.1261.11
RS-CPTBackbone=Qwen3-14B-Base, CPT Strategy=Random Data Sampling2025.1256.4
LPS-CPTBackbone=Qwen3-14B-Base, CPT Strategy=Low-PPL Sampling2025.1256.4
HPS-CPTBackbone=Qwen3-14B-Base, CPT Strategy=High-PPL Sampling2025.1245.7
Qwen3-14B-BaseBackbone=Qwen3-14B-Base, CPT Strategy=None2025.1241.3