dongboklee

/

gORM-8B-merged

like 0

This PR adds a comprehensive model card for the gORM-8B-merged reward model.

It includes:

• A link to the scientific paper, Rethinking Reward Models for Multi-Domain Test-Time Scaling.
• The Apache-2.0 license, based on common practice and colleague consensus.
• The transformers library compatibility tag, which enables the automated "How to use" widget, as evidenced by config.json and common usage patterns for Llama-based models.
• The text-generation pipeline tag, correctly classifying the model's functionality.
• Relevant descriptive tags (reward-model, reasoning, multi-domain, llama, LLM) for better discoverability.
• A link to the official GitHub repository (https://github.com/db-Lee/Multi-RM).
• A concise sample usage code snippet demonstrating how to use the model with the transformers library, adapted from the GitHub README.
• The BibTeX citation for the paper.
• An overview image from the GitHub repository.

Please review and merge this PR if everything looks good.