Clinical Question Answering on HealthBench (offline)
31.79AccuracySkeMex
Evaluation Results
| Method | Links | |
|---|---|---|
| SkeMexBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 31.79 | |
| GMBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 28.66 | |
| CFMBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 28.28 | |
| SkeMexBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 27.65 | |
| GSEMBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 27.57 | |
| ExPeLBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 27.56 | |
| MobileEBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 27.5 | |
| SkillWeaverBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 27.36 | |
| Agent KBBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 27.23 | |
| ReflexionBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 26.89 | |
| DILUBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 26.89 | |
| MempBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 26.74 | |
| AWMBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 26.6 | |
| CRITICBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 26.29 | |
| GSEMBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 26 | |
| EvolverBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 25.85 | |
| VoyagerBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 25.44 | |
| DCBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 25.06 | |
| ReflexionBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 24.54 | |
| ReActBackbone=Qwen3.6-Plus, Setting=Offline2026.06 | 23.98 | |
| DCBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 23.76 | |
| VoyagerBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 23.74 | |
| ExPeLBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 23.63 | |
| GMBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 23.49 | |
| SkillWeaverBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 23.42 | |
| DILUBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 23.37 | |
| CFMBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 23.26 | |
| Agent KBBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 23.08 | |
| AWMBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 22.84 | |
| CoTBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 22.42 | |
| MobileEBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 21.97 | |
| EvolverBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 21.59 | |
| CRITICBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 21.53 | |
| MempBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 21.44 | |
| ReActBackbone=DeepSeek-V3.2, Setting=Offline2026.06 | 19.06 | |
| CoTBackbone=MedGemma-27B, Setting=Offline2026.06 | 14.83 | |
| CoTBackbone=HuluMed-32B, Setting=Offline2026.06 | 9.11 | |
| CoTBackbone=Lingshu-32B, Setting=Offline2026.06 | 8.88 |