Diagnostic Precision on MIMIC-IV (test)
57.7AccuracyMedCollab
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| MedCollab2026.03 | 57.7 | 48.3 | 37.8 | 83.7 | 100 | |
| Gemini-3-FlashCategory=Leading LLMs2026.03 | 55 | 41.8 | 36.3 | 71.6 | 100 | |
| Qwen3-MaxCategory=Leading LLMs2026.03 | 54.1 | 37.1 | 31.7 | 64 | 100 | |
| GPT-4oCategory=Leading LLMs2026.03 | 53.4 | 36.6 | 33.7 | 64.9 | 100 | |
| GLM-4.7Category=Leading LLMs2026.03 | 51.6 | 37.6 | 32.8 | 68.4 | 100 | |
| ClinicalAgentCategory=Multi-Agents2026.03 | 51.6 | 29.2 | 28.8 | 54.3 | 100 | |
| MedLACategory=Multi-Agents2026.03 | 49.6 | 35 | 35.8 | 64.7 | 100 | |
| Baichuan4-TurboCategory=Leading LLMs2026.03 | 47.1 | 29.1 | 39.7 | 59.7 | 98 | |
| Qwen3-8BCategory=Leading LLMs2026.03 | 44.9 | 28.6 | 37.8 | 65.5 | 98.1 | |
| HuatuoGPT-o1-8BCategory=Leading LLMs2026.03 | 32.6 | 23.9 | 26.6 | 59.8 | 82.5 | |
| MEDDxAgentCategory=Multi-Agents2026.03 | 31.8 | — | 26 | — | — |