KID-7391

/

CoTAP

like 0

This PR enhances the model card for CoTAP: Semantic Concentration for Self-Supervised Dense Representations Learning.

Key improvements include:

• Adding the pipeline_tag: image-feature-extraction to improve discoverability on the Hugging Face Hub (e.g., at https://huggingface.co/models?pipeline_tag=image-feature-extraction).
• Including library_name: pytorch as the model's code is based on PyTorch, enabling automated usage snippets.
• Integrating the full paper abstract to provide a comprehensive overview of the research.
• Linking to the official paper: https://huggingface.co/papers/2509.09429.
• Linking to the official GitHub repository: https://github.com/KID-7391/CoTAP.
• Incorporating detailed sections from the GitHub README, such as Checkpoints, Environments, Installation, Training, and Evaluation, to provide clear instructions for users.
• Embedding the pipeline overview image from the GitHub repository.

These additions make the model card more informative, accessible, and aligned with Hugging Face Hub standards.