MedVLSynther

/

MedVLSynther-3B-RL_5K_glm-glm

like 0

Follow

MedVLSynther 3

This PR enhances the model card by adding key metadata and improving content structure:

• pipeline_tag: image-text-to-text: This ensures the model is discoverable under multimodal VQA tasks, aligning with its functionality of performing "joint reasoning over images and text" for medical VQA.
• library_name: transformers: The provided usage snippet explicitly uses classes from the transformers library, confirming compatibility and enabling the automated "How to use" widget on the model page.
• A direct link to the paper, MedVLSynther: Synthesizing High-Quality Visual Question Answering from Medical Documents with Generator-Verifier LMMs, has been added to the top of the model card content for improved accessibility and context.

All existing and accurate information, including the code example, project page link, and detailed model description, has been preserved.