hzf666

/

TSPO-0.4B

like 0

This PR significantly enhances the model card for TSPO by:

• Activating the existing content, which was previously commented out.
• Adding license: apache-2.0, pipeline_tag: video-text-to-text, and library_name: transformers to the YAML metadata, which improves discoverability and provides crucial information at a glance.
• Including descriptive tags: video-understanding, reinforcement-learning, and long-video.
• Updating the content to match the more comprehensive GitHub README.
• Replacing the arXiv paper link with the official Hugging Face paper page: TSPO: Temporal Sampling Policy Optimization for Long-form Video Language Understanding.
• Adding a link to the project page: https://vision-cair.github.io/LongVU.
• Correcting image paths to ensure they render correctly on the Hugging Face Hub.
• Adding a proper BibTeX citation for the paper.

This makes the model more accessible and informative for researchers and practitioners on the Hugging Face Hub.