smz8599

/

GUI-AIMA-lite-3B

like 1

This PR adds a comprehensive model card for the GUI-AIMA-3B model, as presented in the paper GUI-AIMA: Aligning Intrinsic Multimodal Attention with a Context Anchor for GUI Grounding.

It includes:

• Essential metadata: license: cc-by-nc-4.0, pipeline_tag: image-text-to-text for discoverability, and library_name: transformers for automated code snippet generation, based on compatibility evidence.
• Direct links to the paper, the project page, and the GitHub repository.
• A concise summary of the model from the paper's abstract, alongside key architectural and result images from the GitHub README.
• The detailed "Main Results" section, including associated images, to highlight the model's performance.
• A "Sample Usage" code snippet, extracted from eval/example_inference.py in the official GitHub repository, to help users get started easily.
• The relevant BibTeX citation.

This model card significantly improves the documentation and usability of the GUI-AIMA-3B model on the Hugging Face Hub.

Please review and merge this PR if everything looks good.