High-level Diagnostic Reasoning on EndoAgentBench
76.03CAP (CAR)GPT-4o
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| GPT-4oCategory=General MLLMs2025.08 | 76.03 | 50.19 | 63.1 | |
| EndoCogniAgentCategory=Our Method2025.08 | 64 | 78.24 | 71.13 | |
| Step-1oCategory=General MLLMs2025.08 | 61.18 | 54.69 | 57.93 | |
| Gemini 2.5 ProCategory=General MLLMs2025.08 | 53.29 | 35.83 | 44.55 | |
| ColonGPTCategory=Medical MLLMs2025.08 | 52.91 | 20.54 | 36.71 | |
| Yi-VisionCategory=General MLLMs2025.08 | 47.09 | 25.52 | 36.29 | |
| HuatuoGPTCategory=Medical MLLMs2025.08 | 41.07 | 20.54 | 30.8 | |
| LLaVA-MedCategory=Medical MLLMs2025.08 | 3.2 | 2.25 | 2.72 |