ChaoyangWang

/

AdaTooler-V-7B

like 0

This PR significantly improves the model card for AdaTooler-V-7B by:

• Adding the pipeline_tag: image-text-to-text metadata, which correctly categorizes the model as a Multimodal Large Language Model processing visual input to generate text responses.
• Specifying library_name: transformers in the metadata. This is supported by evidence in the model's config.json on the Hugging Face Hub, which lists Qwen2VLForConditionalGeneration as an architecture, indicating compatibility with the 📚 Transformers library. This enables the "Use in Transformers" widget on the model page.
• Including a direct link to the official paper: AdaTooler-V: Adaptive Tool-Use for Images and Videos.
• Providing a link to the official GitHub repository: https://github.com/CYWang735/AdaTooler-V.
• Adding a comprehensive description of the model, summarizing its key features, methodology, and performance highlights, based on the paper abstract and the GitHub README.
• Including an illustrative performance image (bar.png) from the GitHub repository to visually summarize the model's capabilities.
• Adding a BibTeX citation for easy referencing.

Please note that a sample usage code snippet has been omitted, as the provided GitHub README does not contain a direct Python inference example for the AdaTooler-V-7B model, in adherence to the task's strict guidelines.