RLinf

/

RLinf-math-7B

like 3

Follow

RLinf 29

This PR enhances the model card for RLinf/RLinf-math-7B by:

• Updating the pipeline_tag to text-generation to accurately reflect the model's primary inference capability as a causal language model for mathematical reasoning, as indicated by config.json and configuration.json.
• Adding reinforcement-learning to the tags list to preserve information about the model's training methodology.
• Adding library_name: transformers to the metadata, as demonstrated by the existing How to Use section. This will enable the automatic code snippet on the Hugging Face Hub.
• Prominently linking the paper RLinf-VLA: A Unified and Efficient Framework for VLA+RL Training at the top of the model card.

These changes will improve the model's discoverability and provide more complete information for users.