FlyLee

/

bayesian-peft

like 1

This PR significantly enhances the model card by:

• Adding pipeline_tag: text-generation to ensure discoverability on the Hub for LLM tasks.
• Specifying library_name: transformers as the model is compatible with the 🤗 Transformers library, enabling the interactive "How to use" widget.
• Including a detailed description of the model based on the paper's abstract, along with a direct link to the research paper: Training-Free Bayesianization for Low-Rank Adapters of Large Language Models.
• Providing a link to the official GitHub repository (https://github.com/Wang-ML-Lab/bayesian-peft) for comprehensive code and usage instructions.
• Adding a "Usage" section that points users to the explicit bash scripts and LoRA loading commands found in the GitHub repository, in adherence to the guidelines about not generating code snippets without direct evidence.

These additions will make the model more accessible, discoverable, and easier for the community to understand and utilize.