RobinWZQ

/

poisoned_model_1

like 0

This PR enriches the model card for the model associated with the paper Assimilation Matters: Model-level Backdoor Detection in Vision-Language Pretrained Models.

It adds the following:

• library_name: transformers metadata, enabling the "how to use" widget, as the config.json indicates compatibility with transformers.
• pipeline_tag: zero-shot-image-classification metadata, appropriate for a CLIPTextModel component within a VLP context.
• A link to the GitHub repository.
• A summary of the paper's contribution (AMDET).
• Authors of the paper.
• A sample usage code snippet for backdoor detection, directly from the GitHub README.
• The BibTeX citation placeholder from the GitHub README.

Please review and merge if it looks good.