Improve model card for Variational Reasoning for Language Models

#1
by nielsr HF Staff - opened

This PR updates the model card for the Variational Reasoning framework.

It adds:

  • Relevant pipeline_tag: text-generation and tags: [reasoning, qwen] to the YAML metadata for improved discoverability.
  • A comprehensive Model Description derived from the paper's abstract and GitHub README.
  • Direct links to the official paper (https://huggingface.co/papers/2509.22637) and the GitHub repository (https://github.com/sail-sg/variational-reasoning).
  • Details about model development, type, and language.
  • Guidance for Direct Use and Out-of-Scope Use.
  • Information regarding Training Details and Evaluation, linking back to the GitHub repository for in-depth procedures.
  • The official BibTeX citation.

This update ensures the model card is informative and adheres to documentation best practices, without including speculative information such as a license or made-up usage examples.

Ready to merge
This branch is ready to get merged automatically.

Sign up or log in to comment