xxwu

/

Agent-STAR-RL-7B

like 0

Hi! I'm Niels from the Hugging Face community science team. I've opened this PR to improve the documentation for Agent-STAR-RL-7B.

This PR adds:

• A link to the associated paper: Demystifying Reinforcement Learning for Long-Horizon Tool-Using Agents: A Comprehensive Recipe.
• A link to the official GitHub repository.
• Metadata including pipeline_tag, library_name, and base_model.
• A detailed model description and the official BibTeX citation.

These additions help users better understand the model's origin, its training recipe (the STAR pipeline), and how to use it correctly in agentic environments.