List your company

Medical Question Answering on MedQA, MedMCQA, and PubMedQA (Aggregate 1,000 each)

63.67AccuracyProposed
-2.47414.69831.8749.042May 31, 2026

Evaluation results

MethodLinks
ProposedModel=Qwen2.5-7B2026.0563.6710099.9710.26
CoTModel=Qwen2.5-7B2026.0560.0394.40.284.78
ProposedModel=Phi-4-mini2026.0557.6799.1399.240.54
DirectModel=Qwen2.5-7B2026.0556.591.470.0792.55
HypMed-v3Model=Qwen2.5-7B2026.0556.2398.699.979.73
HypMed-v3Model=Phi-4-mini2026.0549.4393.199.243.41
CoTModel=Phi-4-mini2026.0548.1779.23098.38
ProposedModel=DeepSeek-R1-32B2026.0538.7772.1344.32.13
HypMed-v3Model=DeepSeek-R1-32B2026.0534.1357.9344.32.66
DirectModel=Phi-4-mini2026.0530.254.03083.29
DirectModel=DeepSeek-R1-32B2026.0516.935.570.129.86
ProposedModel=BioMistral-7B2026.0513.0733.30.40
CoTModel=BioMistral-7B2026.0512.4732.03082.87
CoTModel=DeepSeek-R1-32B2026.057.123.70.10
DirectModel=BioMistral-7B2026.050.81.47013.1
HypMed-v3Model=BioMistral-7B2026.050.070.40.40