junyoung-00

/

Phi-3.5-vision-instruct-ChartCap

like 4

This PR significantly enhances the model card for the ChartCap model by:

• Adding essential metadata: pipeline_tag: image-to-text and library_name: transformers for improved discoverability and integration with Hugging Face Hub features.
• Including relevant tags to further categorize the model (chart-captioning, multimodal, vision-language-model).
• Providing direct links to the official paper: ChartCap: Mitigating Hallucination of Dense Chart Captioning.
• Adding the project page link: https://junyoung-00.github.io/ChartCap/.
• Including the GitHub repository link: https://github.com/junyoung-00/ChartCap.
• Adding a comprehensive model description.
• Providing a practical Python sample usage snippet to demonstrate how to load and use the model for chart captioning.
• Adding a section on the ChartCap dataset.
• Including BibTeX citation information.

These updates will make the model more informative, discoverable, and user-friendly for the community.