Gepardius

/

CLICITA

Like 0

This PR significantly enhances the model card for CLICITA: Continual Learning for Image Captioning through Improved Image-Text Alignment by:

• Adding pipeline_tag: image-to-text to enable discoverability on the Hugging Face Hub (e.g., at https://huggingface.co/models?pipeline_tag=image-to-text).
• Adding library_name: transformers as the model uses a ViT-GPT-2 backbone and the project relies on the transformers library (evidenced by transformers in requirements.txt), allowing for an automated "how to use" widget.
• Adding continual-learning as an additional tag, which is highly relevant to the paper's core contribution.
• Providing a clear link to the official paper on Hugging Face: Continual Learning for Image Captioning through Improved Image-Text Alignment.
• Adding a clear link to the GitHub repository for easy access to the code: https://github.com/Gepardius/Taetz_Bordelius_Continual_ImageCaptioning.
• Incorporating the comprehensive description, method overview, results, installation, usage examples, and other details directly from the GitHub README to provide a complete overview of the model.
• Correcting the placeholder <your-org> to Gepardius in the installation commands.

Please review and merge if everything looks good!