SmoothLLM: Defending Large Language Models Against Jailbreaking Attacks

Authors: Alexander Robey, Eric Wong, Hamed Hassani, George J. Pappas
Published: 2023-10-05 | Updated: 2024-06-11

Source: https://arxiv.org/abs/2310.03684

PDF: https://arxiv.org/pdf/2310.03684

Labels Predicted by AI

防御手法 プロンプトインジェクション LLM性能評価

Please note that these labels were automatically added by AI. Therefore, they may not be entirely accurate.
For more details, please see the About the Literature Database page.

Abstract

Despite efforts to align large language models (LLMs) with human intentions, widely-used LLMs such as GPT, Llama, and Claude are susceptible to jailbreaking attacks, wherein an adversary fools a targeted LLM into generating objectionable content. To address this vulnerability, we propose SmoothLLM, the first algorithm designed to mitigate jailbreaking attacks. Based on our finding that adversarially-generated prompts are brittle to character-level changes, our defense randomly perturbs multiple copies of a given input prompt, and then aggregates the corresponding predictions to detect adversarial inputs. Across a range of popular LLMs, SmoothLLM sets the state-of-the-art for robustness against the GCG, PAIR, RandomSearch, and AmpleGCG jailbreaks. SmoothLLM is also resistant against adaptive GCG attacks, exhibits a small, though non-negligible trade-off between robustness and nominal performance, and is compatible with any LLM. Our code is publicly available at https://github.com/arobey1/smooth-llm.