nyu-visionx

/

RAE-collections

like 41

Follow

VISIONx @ NYU 141

This PR improves the model card for the "MeanFlow Transformers with Representation Autoencoders" model by:

• Adding the pipeline_tag: unconditional-image-generation to the metadata, enhancing discoverability for image generation tasks.
• Including a direct link to the official GitHub repository (https://github.com/sony/mf-rae) for easy access to the codebase.
• Providing a concise overview of the model and its key achievements from the paper abstract to give users a better understanding of its capabilities.

The existing arXiv paper link and project website link have been retained as per instructions. A library_name tag and a sample usage section were omitted due to lack of explicit code evidence in the GitHub README that meets the documentation guidelines.

Please review these additions.