Year: 2024
Type: preprint
Abstract: Adversarial misuse, particularly through `jailbreaking' that circumvents a model's safety and ethical protocols, poses a significant challenge for Large Language Models (LLMs). This paper delves into the mechanisms behind such successful attacks, introducing a hypothesis for the safety mechanism of aligned LLMs: intent security recognition followed by response generation. Grounded in this hypothesis, we propose CodeChameleon, a novel jailbreak framework based on personalized encryption tactics. To elude the intent security recognition phase, we reformulate tasks into a code completion format, enabling users to encrypt queries using personalized encryption functions. To guarantee response generation functionality, we embed a decryption function within the instructions, which allows the LLM to decrypt and execute the encrypted queries successfully. We conduct extensive experiments on 7 LLMs, achieving state-of-the-art average Attack Success Rate (ASR). Remarkably, our method achieves an 86.6\% ASR on GPT-4-1106.
Source: arXiv (Cornell University)
Authors: Lv, Huijie, Wang, Xiao, Zhang, Yuansen, Huang, Caishuang, Dou, Shihan, Ye, Junjie, Gui, Tao, Zhang, Qi, Huang, Xuanjing
Institution: Fudan University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4392222991