Open-set diagnostic naming on DDXPlus
48.5AccuracyCLR-voyance-8B
Evaluation Results
| Method | Links | |
|---|---|---|
| CLR-voyance-8BFamily=8B family, Merge Strategy=DELLA-Linear2026.05 | 48.5 | |
| GPT-5Family=Frontier API2026.05 | 46.2 | |
| HuatuoGPT-o1Family=Medical / open-source2026.05 | 46 | |
| Qwen3-8B baseFamily=8B family2026.05 | 43.5 | |
| GPT-4.1Family=Frontier API2026.05 | 43 | |
| GPT-4.1-miniFamily=Frontier API2026.05 | 42 | |
| MG-4B baseFamily=4B family2026.05 | 37.5 | |
| Gemma-3-4BFamily=Medical / open-source2026.05 | 36 | |
| CLR-voyance-4BFamily=4B family, Merge Strategy=DELLA-Linear2026.05 | 27.5 | |
| MG-Δ+AFamily=4B family, Merge Strategy=DELLA-Linear+AIM2026.05 | 26 | |
| MG-BFamily=4B family, Merge Strategy=Breadcrumbs2026.05 | 25.5 | |
| MG-B+AFamily=4B family, Merge Strategy=Breadcrumbs+AIM2026.05 | 24.5 | |
| DS-R1-0528Family=Medical / open-source2026.05 | 18.5 | |
| DS-R1-DistillFamily=Medical / open-source2026.05 | 15 | |
| MedGemma-27BFamily=Medical / open-source2026.05 | 8 |