yuxiaw
commented
3 months ago Hi Xingquan,
Thanks for your question, the code based on LLaMA-3 was sorted recently to make it more readable and clean, but we still did not push to the public github repository. If you need them urgently, I can send them to your personal email, while the clean version will be coming in the following month.
Best, Yuxia
On Sat, Jun 1, 2024 at 1:21 PM Zhang Xingquan @.***> wrote:
Hello, I am working on a fact-checking system and would like to use the LLAMA-3 8B model as a verifier. I came across your paper that describes this approach and provides code, but I am encountering difficulties in modifying the code to support the LLAMA-3 model and reproduce the results described in the paper. I would appreciate any guidance and support from the community. Here are my questions:
- What specific modifications are needed in the provided code to integrate the LLAMA-3 8B model as a verifier?
- Are there any detailed steps or examples available for adapting the code to work with LLAMA-3 8B?
Thank you very much for your help!
— Reply to this email directly, view it on GitHub https://github.com/yuxiaw/OpenFactCheck/issues/2, or unsubscribe https://github.com/notifications/unsubscribe-auth/APIGIX7PLDDYOQTEYP7E3QTZFGHBZAVCNFSM6AAAAABIT76XXSVHI2DSMVQWIX3LMV43ASLTON2WKOZSGMZDSMBTGE4DIMI . You are receiving this because you are subscribed to this thread.Message ID: @.***>
Hello, I am working on a fact-checking system and would like to use the LLAMA-3 8B model as a verifier. I came across your paper that describes this approach and provides code, but I am encountering difficulties in modifying the code to support the LLAMA-3 model and reproduce the results described in the paper. I would appreciate any guidance and support from the community. Here are my questions:
Thank you very much for your help!