MedVLSynther

/

MedVLSynther-3B-RL_5K_qwen-glm

like 0

Follow

MedVLSynther 3

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

• pipeline_tag: image-text-to-text: This tag correctly categorizes the model for multimodal Visual Question Answering (VQA) tasks, significantly improving its discoverability on the Hugging Face Hub.
• library_name: transformers: Based on the config.json, tokenizer_config.json, and the provided usage example, the model is fully compatible with the transformers library. Adding this tag will enable the automated "how to use" widget on the model page, providing users with ready-to-run code snippets.
• Paper Link: An explicit link to the Hugging Face paper page (MedVLSynther: Synthesizing High-Quality Visual Question Answering from Medical Documents with Generator-Verifier LMMs) has been added at the top for easy access to the research.
• Redundant License: The duplicate "License" section at the end of the model card has been removed, as the license is already clearly defined in the YAML metadata, maintaining consistency and avoiding repetition.

These updates collectively improve the model card's completeness, usability, and adherence to Hub best practices.