Close-ended VQA on NEJM Image Challenge
69.7AccuracyMed-Gemini-L 1.0
Evaluation results
| Method | Links | |
|---|---|---|
| Med-Gemini-L 1.0Multimodal fine-tuned=false2024.04 | 69.7 | |
| GPT-4VMultimodal fine-tuned=false2024.04 | 61 |
| Method | Links | |
|---|---|---|
| Med-Gemini-L 1.0Multimodal fine-tuned=false2024.04 | 69.7 | |
| GPT-4VMultimodal fine-tuned=false2024.04 | 61 |