sinaalemohammad

/

Neon

like 0

This PR significantly enhances the model card for Neon: Negative Extrapolation From Self-Training, a method for improving generative AI models.

It adds:

• The pipeline_tag: unconditional-image-generation to ensure proper discoverability on the Hugging Face Hub.
• A comprehensive description of the model, including its abstract, key methodology, and performance benchmarks.
• Links to the official paper: Neon: Negative Extrapolation From Self-Training Improves Image Generation.
• A link to the official GitHub repository: https://github.com/SinaAlemohammad/Neon.
• Details on the method, benchmark performance, repository map, citation, contact information, and acknowledgments, all extracted from the official GitHub README.
• A reference to the GitHub repository for detailed quickstart and evaluation instructions, adhering to the "no made-up code" policy for sample usage.

The license remains mit as found in the existing model card, and library_name is intentionally omitted due to the lack of direct Hugging Face library compatibility for automated inference snippets.