Medical Visual Question Answering on DeepTumorVQA refined 2025b
66Total Average3DMedAgent (GPT-5)
Evaluation results
| Method | Links | |
|---|---|---|
| 3DMedAgent (GPT-5)Base Model=GPT-5, Evaluation Protocol=zero-shot2026.02 | 66 | |
| 3DMedAgent (Qwen3-VL)Base Model=Qwen3-VL (30B), Evaluation Protocol=zero-shot2026.02 | 61 | |
| Qwen3-VLParameters=30B, Evaluation Protocol=zero-shot2026.02 | 42 | |
| GPT-5Evaluation Protocol=zero-shot2026.02 | 41 | |
| M3DEvaluation Protocol=zero-shot2026.02 | 39 | |
| MedGemmaParameters=27B, Evaluation Protocol=zero-shot2026.02 | 38 | |
| HuatuoGPTParameters=34B, Evaluation Protocol=zero-shot2026.02 | 38 | |
| RandType=Random baseline2026.02 | 37 | |
| RadFMEvaluation Protocol=zero-shot2026.02 | 37 |