TruthX
TruthX copied to clipboard
Code for ACL 2024 paper "TruthX: Alleviating Hallucinations by Editing Large Language Models in Truthful Space"
I used code from your example and `generate` method of llama2-chat-7B-TruthX produces different output than base llama2-chat-7B. However I have a problem when I try to extract the logits (and...
Great work! Do you have any plan to release the training code for the TruthX model?
Hello! Thanks for your excellent work! It seems like this repo does not provide the code for training TruthX. Could you release the code for training TruthX? Looking forward to...
### **Issue Description** I followed the instructions in the repository to reproduce the TruthX evaluation on **Llama-2-7B-Chat** but observed significant discrepancies in **multiple-choice metrics** and **generation outputs** compared to the...
Could you provide the details of the generalization experiment in Table 2? When conducting generalization experiments on the Natural Questions, TriviaQA, and FACTOR datasets, was the data used directly from...
I'd like to try more baked-in models like qwen2-7B and baichuan2-7B. Are you interested in releasing more?
Hi, thanks for sharing your work! I’m having some trouble reproducing the results and would appreciate your help. For the TruthfulQA Multiple-Choice task, I got these results when evaluating LLMs...
当我从huggingface上下载模型llama2-7b-chat-truthX后 运行text.py时会出现如下错误,请问该如何解决?