List your company

Question Answering on MedMCQA (R@1, R@3, N@1, N@3)

72.33R@1S-DPO
22.441235.393148.34561.2969May 11, 2026

Evaluation results

MethodLinks
S-DPOModel=Llama32026.0572.3394.3472.3385.2
MASS-DPOModel=Llama32026.0571.2394.4971.2384.84
DPO-kModel=Llama32026.0571.1393.8871.1384.51
MASS-DPOModel=Qwen32026.0556.3489.7256.3475.62
DPO-kModel=Qwen32026.0554.5989.6754.5974.86
DPOModel=Llama32026.0551.4887.8251.4872.3
S-DPOModel=Qwen32026.0551.0386.5251.0371.54
S-DPOModel=SmolLM32026.0545.4687.2245.4669.54
MASS-DPOModel=SmolLM32026.0544.8187.4744.8169.4
DPO-kModel=SmolLM32026.0544.1685.9144.1668.09
DPOModel=Qwen32026.0539.2584.5139.2565.37
DPOModel=SmolLM32026.0533.3381.7533.3361.01
DMPOModel=Qwen32026.0526.0274.8926.0253.6
DMPOModel=SmolLM32026.0526.0275.3926.0253.93
DMPOModel=Llama32026.0524.3674.9924.3652.84