Adversarial Tuning: Defending Against Jailbreak Attacks for LLMs

Abstract

Although safely enhanced Large Language Models (LLMs) have achieved remarkable success in tackling various complex tasks in a zero-shot manner, they remain susceptible to jailbreak attacks, particularly the unknown jailbreak attack. To enhance LLMs' generalized defense capabilities, we propose a two-stage adversarial tuning framework, which generates adversarial prompts to explore worst-case scenarios by optimizing datasets containing pairs of adversarial prompts and their safe responses. In the

Categories

Framework mappings

Suggested from the entry's categories.

@misc{liu2024adversarial,
  title = {{Adversarial Tuning: Defending Against Jailbreak Attacks for LLMs}},
  author = {Fan Liu and Zhao Xu and Hao Liu},
  year = {2024},
  month = jun,
  eprint = {2406.06622},
  archivePrefix = {arXiv},
  doi = {10.48550/arXiv.2406.06622},
  url = {https://www.semanticscholar.org/paper/9223a64feb573e62498c2ca914ed97557c580167}
}