OX-PIXL

/

SpatialThinker-3B

like 2

Follow

Perceptual Intelligence and Extended Reality Lab 6

This PR significantly enhances the model card for SpatialThinker by:

• Adding the pipeline_tag: image-text-to-text to improve discoverability for multimodal vision-language tasks.
• Adding library_name: transformers as evidenced by the model's architecture files and GitHub requirements, which will enable the automated "how to use" widget.
• Expanding the model description with the paper's abstract.
• Including direct links to the Hugging Face paper page, the project page, and the GitHub repository.
• Integrating key sections from the GitHub README, including Requirements, Installation, Training, Merge Checkpoints, Evaluation (with shell command examples), Supported Evaluation Datasets, Citation (BibTeX), and Acknowledgements to provide comprehensive usage information.
• Adding the overview image from the GitHub repository.

These updates aim to provide a more complete and useful model card for the Hugging Face community.