Hugging Face's logo

Kwai-Klear
/
Klear-Reasoner-8B

Safetensors
English
qwen3
Model card Files
xet
Community
3

Improve model card: Add pipeline tag, library name, and update CE-GPPO paper/code links

#3
by nielsr HF Staff - opened
base: refs/heads/main
←
from: refs/pr/3
Discussion Files changed
+27
-20
nielsr

This PR enhances the model card for Kwai-Klear/Klear-Reasoner-8B, which is based on the CE-GPPO method.

Key improvements include:

  • Adding pipeline_tag: text-generation to improve discoverability on the Hugging Face Hub.
  • Specifying library_name: transformers to enable the automated "how to use" widget, as the model is compatible with the transformers library.
  • Updating the primary paper links in the model card to point to the CE-GPPO paper (CE-GPPO: Controlling Entropy via Gradient-Preserving Clipping Policy Optimization in Reinforcement Learning).
  • Updating the GitHub link to the official CE-GPPO repository (https://github.com/Kwai-Klear/CE-GPPO).
  • Updating the "Training" section to reflect the instructions from the CE-GPPO GitHub repository for better consistency and accuracy.
  • Updating the citation section to reference the correct CE-GPPO paper.

Please review and merge this PR if everything looks good.

 Improve model card: Add pipeline tag, library name, and update CE-GPPO paper/code linksf5548f1f
Ready to merge
This branch is ready to get merged automatically.

Sign up or log in to comment