MMR1

/

MMR1-3B-RL

like 0

Follow

MMR1 13

This Pull Request significantly enhances the model card for MMR1: Enhancing Multimodal Reasoning with Variance-Aware Sampling and Open Resources.

Key improvements include:

• Adding license: apache-2.0, library_name: transformers, and pipeline_tag: image-text-to-text to the metadata for better discoverability and Hub integration.
• Including descriptive tags: multimodal-llm, qwen, reasoning, llm.
• Incorporating the paper's abstract for a comprehensive overview of the model.
• Expanding the content section with detailed explanations of the model's introduction, methodology, open resources, evaluation results, and qualitative demos, directly sourced from the project's GitHub README.
• Ensuring all images are correctly linked using raw GitHub URLs and internal links are robust.
• Adding explicit links to the paper, GitHub repository, and related Hugging Face assets (models, datasets).

These updates aim to make the model card more informative, discoverable, and user-friendly for the community.