QiWang98

/

VideoRFT-SFT-3B

like 0

This PR significantly enhances the model card for VideoRFT, linking it to the paper VideoRFT: Incentivizing Video Reasoning Capability in MLLMs via Reinforced Fine-Tuning.

Key updates include:

• Adding library_name: transformers to enable the "Use in Transformers" widget, as the model is fully compatible with the library.
• Updating pipeline_tag from visual-question-answering to video-text-to-text to better reflect the model's capabilities in video reasoning and understanding.
• Populating the main content section with a detailed overview (including the abstract), methodology, dataset information, installation/training/evaluation instructions, and a runnable "Quick Inference Code" example, all extracted from the project's GitHub README.
• Ensuring all relevant links (paper, GitHub, datasets) are correctly included.

This comprehensive update aims to improve discoverability and usability for researchers and developers on the Hugging Face Hub.