List your company

Medical Domain Question Answering on HealthBench In-domain, Seen

22.97ScoreEvoRubrics
5.2389.841514.44519.0485Jun 22, 2026

Evaluation results

MethodLinks
EvoRubricsModel=Qwen3-8B, Category=Ours2026.0622.970.4227
EvoRubricsModel=Qwen3-4B, Category=Ours2026.0620.80.3664
GoldenRubricsModel=Qwen3-8B, Category=RL w/ Rubric2026.0620.470.3833
OnlineRubricsModel=Qwen3-8B, Category=RL w/ Rubric2026.0620.230.3774
SFTModel=Qwen3-8B, Category=Supervised2026.0618.520.3427
PromptModel=Deepseek-V4-Flash, Category=Vanilla2026.0618.040.3303
PromptModel=Qwen3-32B, Category=Vanilla2026.0617.170.3207
GoldenRubricsModel=Qwen3-4B, Category=RL w/ Rubric2026.0615.820.2775
PromptModel=Qwen3-8B, Category=Vanilla2026.0614.40.2592
SFTModel=Qwen3-4B, Category=Supervised2026.0613.040.2375
PromptModel=Qwen3-4B, Category=Vanilla2026.0612.620.2248
RuscaRLModel=Qwen3-4B, Category=RL w/ Rubric2026.068.850.1889
RuscaRLModel=Qwen3-8B, Category=RL w/ Rubric2026.068.240.1423
OnlineRubricsModel=Qwen3-4B, Category=RL w/ Rubric2026.065.920.0804