List your company

Medical Domain Question Answering on RaR-Medicine In-domain Unseen

20.22ScorePrompt
10.090412.720215.3517.9798Jun 22, 2026

Evaluation results

MethodLinks
PromptModel=Deepseek-V4-Flash, Category=Vanilla2026.0620.2275
PromptModel=Qwen3-32B, Category=Vanilla2026.0620.0574.69
EvoRubricsModel=Qwen3-8B, Category=Ours2026.0617.3461.71
PromptModel=Qwen3-8B, Category=Vanilla2026.061760.51
EvoRubricsModel=Qwen3-4B, Category=Ours2026.0616.3660.5
OnlineRubricsModel=Qwen3-8B, Category=RL w/ Rubric2026.0616.3257.85
GoldenRubricsModel=Qwen3-8B, Category=RL w/ Rubric2026.0616.1657.26
SFTModel=Qwen3-8B, Category=Supervised2026.0615.9456.5
PromptModel=Qwen3-4B, Category=Vanilla2026.0615.1456.04
RuscaRLModel=Qwen3-8B, Category=RL w/ Rubric2026.0612.6744.82
SFTModel=Qwen3-4B, Category=Supervised2026.0612.3545.67
GoldenRubricsModel=Qwen3-4B, Category=RL w/ Rubric2026.0611.4142.16
RuscaRLModel=Qwen3-4B, Category=RL w/ Rubric2026.0610.9840.94
OnlineRubricsModel=Qwen3-4B, Category=RL w/ Rubric2026.0610.4838.66