QiWang98

/

VideoRFT-SFT

like 0

This PR significantly enhances the model card by:

• Updating the pipeline_tag from visual-question-answering to video-text-to-text to better reflect the model's comprehensive video reasoning capabilities.
• Adding library_name: transformers to enable the automated "Use in Transformers" widget, as the model's configuration files and GitHub requirements demonstrate compatibility with the library.
• Populating the content with the paper's abstract (as "Overview"), methodology, dataset details, setup instructions, training and evaluation guides, and acknowledgements, all sourced directly from the project's GitHub README.
• Ensuring that the official paper link and GitHub repository link are prominently displayed.
• Carefully linking all images to their raw GitHub URLs for proper rendering on the Hub.

This update provides a much richer and more accurate overview for users, improving discoverability and ease of use.