List your company

Multi-Turn Medical Dialogue on MedMCQA

63.31AccuracyMEDIQ
19.390830.792942.19553.5971Feb 10, 2026

Evaluation results

MethodLinks
MEDIQModel=Gemini-2.5-Pro, Method Type=Prompt2026.0263.31
TreePOModel=Qwen3-8B, Method Type=SFT+RL2026.0254.74
ATPO (U1 + U2)Model=Qwen3-8B, Method Type=SFT+RL2026.0253.66
ATPO (U1)Model=Qwen3-8B, Method Type=SFT+RL2026.0253.22
MEDIQModel=GPT-4o, Method Type=Prompt2026.0253.03
PPO (H-MDP)Model=Qwen3-8B, Method Type=SFT+RL2026.0251.98
GRPOModel=Qwen3-8B, Method Type=SFT+RL2026.0251.12
PPO (MDP)Model=Qwen3-8B, Method Type=SFT+RL2026.0250
SFTModel=Qwen3-8B, Method Type=SFT2026.0246.87
ATPO (U1)Model=Qwen3-4B, Method Type=SFT+RL2026.0246.23
DirectModel=Qwen3-8B, Method Type=Prompt2026.0246.16
ATPO (U1 + U2)Model=Qwen3-4B, Method Type=SFT+RL2026.0245.93
TreePOModel=Qwen3-4B, Method Type=SFT+RL2026.0245.22
GRPOModel=Qwen3-4B, Method Type=SFT+RL2026.0243.84
PPO (H-MDP)Model=Qwen3-4B, Method Type=SFT+RL2026.0243.32
PPO (MDP)Model=Qwen3-4B, Method Type=SFT+RL2026.0242.5
DFTModel=Qwen3-8B, Method Type=SFT2026.0242.2
MEDIQModel=Qwen3-8B, Method Type=Prompt2026.0241.6
DirectModel=Qwen3-4B, Method Type=Prompt2026.0241.53
ATPO (U1 + U2)Model=Qwen3-1.7B, Method Type=SFT+RL2026.0239.93
SFTModel=Qwen3-4B, Method Type=SFT2026.0239.18
ATPO (U1)Model=Qwen3-1.7B, Method Type=SFT+RL2026.0238.66
TreePOModel=Qwen3-1.7B, Method Type=SFT+RL2026.0238.47
GRPOModel=Qwen3-1.7B, Method Type=SFT+RL2026.0236.57
PPO (MDP)Model=Qwen3-1.7B, Method Type=SFT+RL2026.0235.37
PPO (H-MDP)Model=Qwen3-1.7B, Method Type=SFT+RL2026.0234.89
DFTModel=Qwen3-4B, Method Type=SFT2026.0233.28
DirectModel=Qwen3-1.7B, Method Type=Prompt2026.0232.54
MEDIQModel=Qwen3-1.7B, Method Type=Prompt2026.0232.35
MEDIQModel=Qwen3-4B, Method Type=Prompt2026.0231.64
SFTModel=Qwen3-1.7B, Method Type=SFT2026.0228.1
DFTModel=Qwen3-1.7B, Method Type=SFT2026.0221.08