LLaVa-Med
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Open-Ended Medical Visual Chat | LLaVA-Med unseen 50 image-caption pairs 1.0 (evaluation set) | Conversation Score69.73 | 8 | |
| Open-generation vision-language tasks | LLaVA-Med open-generation clients | Average Judge Score9.36 | 3 | |
| Medical Image Answering | LLaVa-Med qa0.2k | mBMR20.04 | 2 | |
| Medical Image Answering | LLaVa-Med qa0.2k (test) | mBMR- | 0 |
Showing 4 of 4 rows