This PR adds a comprehensive model card for the "Image Tokenizer Needs Post-Training" artifact.
Key updates include:
- Addition of `pipeline_tag: image-feature-extraction` to the YAML metadata, making the model more discoverable (e.g., at https://huggingface.co/models?pipeline_tag=image-feature-extraction).
- Inclusion of the paper link: https://huggingface.co/papers/2509.12474.
- Addition of a link to the official project page: https://qiuk2.github.io/works/RobusTok/index.html.
- Addition of a link to the GitHub repository: https://github.com/qiuk2/RobusTok.
- A descriptive overview of the model, its key highlights, and a "Model Zoo" table directly from the GitHub README.
- A visualization to demonstrate the model's improvements.
- The academic citation for the paper.
- Detailed usage instructions are linked to the GitHub repository.
