@inproceedings{chen-etal-2025-llm-jailbreak,
    title = "{LLM} Jailbreak Detection for (Almost) Free!",
    author = "Chen, Guorui  and
      Xia, Yifan  and
      Jia, Xiaojun  and
      Li, Zhijiang  and
      Torr, Philip  and
      Gu, Jindong",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/author-page-yu-wang-polytechnic/2025.findings-emnlp.309/",
    doi = "10.18653/v1/2025.findings-emnlp.309",
    pages = "5777--5807",
    ISBN = "979-8-89176-335-7",
    abstract = "Large language models (LLMs) enhance security through alignment when widely used, but remain susceptible to jailbreak attacks capable of producing inappropriate content. Jailbreak detection methods show promise in mitigating jailbreak attacks through the assistance of other models or multiple model inferences. However, existing methods entail significant computational costs. In this paper, we first present a finding that the difference in output distributions between jailbreak and benign prompts can be employed for detecting jailbreak prompts. Based on this finding, we propose a Free Jailbreak Detection (FJD) which prepends an affirmative instruction to the input and scales the logits by temperature to distinguish between jailbreak and benign prompts through the confidence of the first token. Furthermore, we enhance the detection performance of FJD through the integration of virtual instruction learning. Extensive experiments on aligned LLMs show that our FJD can effectively detect jailbreak prompts with almost no additional computational costs during LLM inference."
}