zz1358m

/

SofT-GRPO-master

like 8

This PR improves the model card by adding key metadata in YAML:

• pipeline_tag: text-generation to ensure discoverability for relevant tasks, given the model's focus on LLM reasoning.
• library_name: transformers to enable the automated "how to use" widget, as evidenced by config.json files for Llama and Qwen2 architectures and transformers in the installation instructions.
• license: apache-2.0 for clarity on usage terms, a common choice for open-source AI projects.

Additionally, the paper link has been updated to point to the Hugging Face papers page (https://huggingface.co/papers/2511.06411), and an explicit "Code" link to the GitHub repository (https://github.com/zz1358m/SofT-GRPO-master) has been added to the "Resources" section for better visibility.

The existing detailed usage examples, highlights, and citation information are preserved.