medcap-pmcoa
The vision encoder is fine-tuned from BiomedCLIP using Meta-Llama-3-8B-Instruct. For more information, please refer to medcap and README_FINETUNE.md.
The model is still in training, and the current version is preliminary.
The vision encoder is fine-tuned from BiomedCLIP using Meta-Llama-3-8B-Instruct. For more information, please refer to medcap and README_FINETUNE.md.
The model is still in training, and the current version is preliminary.