LLM4Binary

/

llm4decompile-22b-v2

like 28

This PR enhances the model card for LLM4Binary/llm4decompile-6.7b-v2 by:

• Adding pipeline_tag: text-generation to improve model discoverability on the Hugging Face Hub, as the model's core task is to decompile binary code into human-readable C source code.
• Adding library_name: transformers based on the explicit usage of AutoTokenizer and AutoModelForCausalLM from the transformers library in the provided sample code, which enables an automated "how to use" snippet.
• Adding a prominent link to the associated paper, Decompile-Bench: Million-Scale Binary-Source Function Pairs for Real-World Binary Decompilation, at the top of the model card.
• Adding a prominent link to the project's GitHub repository at the top of the model card.
• Updating the license text within the content to "MIT and DeepSeek License" to accurately reflect the information provided in the GitHub README.

These updates will provide clearer information for users and better integrate the model within the Hugging Face ecosystem.