Clinical Response Evaluation on Dermatologist clinical agreement dataset (test)
67Agreement - Strongly AgreeGemini 2.5 Pro
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Gemini 2.5 Pro2025.12 | 67 | 25 | 6 | 0.3 | 1.8 | |
| LLM Blender - Ensembleapproach=Ensemble2025.12 | 39.3 | 36.6 | 17.6 | 2.4 | 4.2 | |
| GPT-4o2025.12 | 37.8 | 43.5 | 10.7 | 4.2 | 3.9 | |
| LLaMA 4 Maverick2025.12 | 37.8 | 43.2 | 12.2 | 3.9 | 3 | |
| SkinGPT-42025.12 | 32.1 | 16.1 | 6.8 | 7.1 | 37.8 |