MBZUAI

/

ViMUL

like 0

Follow

Mohamed Bin Zayed University of Artificial Intelligence 685

This PR significantly improves the model card for ViMUL by:

• Adding pipeline_tag: video-text-to-text to ensure better discoverability for video-to-text generation tasks on the Hub.
• Specifying library_name: transformers in the metadata, indicating compatibility with the Hugging Face Transformers library and enabling an automated usage widget.
• Adding descriptive tags: llava, qwen, multilingual, and video-understanding for improved searchability and categorization.
• Including a direct link to the model's GitHub repository (https://github.com/mbzuai-oryx/ViMUL) for easy access to the source code.
• Correcting the metadata structure by moving the license tag to the proper YAML header and removing the redundant `