MK-runner

/

CoGaze

like 1

This PR adds a comprehensive model card for CoGaze, a vision-language pretraining framework for chest X-rays. The model card includes:

• A link to the research paper and official GitHub repository.
• The image-text-to-text pipeline tag for better discoverability.
• An overview of the model's architecture (context- and gaze-guided).
• Citation information.