Re-Triggering Safeguards within LLMs for Jailbreak Detection

Abstract

This paper proposes a jailbreaking prompt detection method for large language models (LLMs) to defend against jailbreak attacks. Although recent LLMs are equipped with built-in safeguards, it remains possible to craft jailbreaking prompts that bypass them. We argue that such jailbreaking prompts are inherently fragile, and thus introduce an embedding disruption method to re-activate the safeguards within LLMs. Unlike previous defense methods that aim to serve as standalone solutions, our approac

Categories

Framework mappings

Suggested from the entry's categories.

@misc{lin2026retriggering,
  title = {{Re-Triggering Safeguards within LLMs for Jailbreak Detection}},
  author = {Zheng Lin and Zhenxing Niu and Haoxuan Ji and Yuzhe Huang and Haichang Gao},
  year = {2026},
  month = may,
  eprint = {2605.10611},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2605.10611}
}