Year: 2023
Type: preprint
Abstract: Recently, ChatGPT has gained significant attention in research due to its ability to interact with humans effectively. The core idea behind this model is reinforcement learning (RL) fine-tuning, a new paradigm that allows language models to align with human preferences, i.e., InstructGPT. In this study, we propose BadGPT, the first backdoor attack against RL fine-tuning in language models. By injecting a backdoor into the reward model, the language model can be compromised during the fine-tuning stage. Our initial experiments on movie reviews, i.e., IMDB, demonstrate that an attacker can manipulate the generated text through BadGPT.
Source: arXiv (Cornell University)
Authors: Shi, Jiawen, Liu, Yixin, Zhou, Pan, Sun, Lichao
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4367000527