ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2407.21659
25
2

Defending Jailbreak Attack in VLMs via Cross-modality Information Detector

31 July 2024
Yue Xu
Xiuyuan Qi
Zhan Qin
Wenjie Wang
    AAML
ArXivPDFHTML
Abstract

Vision Language Models (VLMs) extend the capacity of LLMs to comprehensively understand vision information, achieving remarkable performance in many vision-centric tasks. Despite that, recent studies have shown that these models are susceptible to jailbreak attacks, which refer to an exploitative technique where malicious users can break the safety alignment of the target model and generate misleading and harmful answers. This potential threat is caused by both the inherent vulnerabilities of LLM and the larger attack scope introduced by vision input. To enhance the security of VLMs against jailbreak attacks, researchers have developed various defense techniques. However, these methods either require modifications to the model's internal structure or demand significant computational resources during the inference phase. Multimodal information is a double-edged sword. While it increases the risk of attacks, it also provides additional data that can enhance safeguards. Inspired by this, we propose C‾\underline{\textbf{C}}C​ross-modality I‾\underline{\textbf{I}}I​nformation DE‾\underline{\textbf{DE}}DE​tectoR‾\underline{\textbf{R}}R​ (CIDER)\textit{CIDER})CIDER), a plug-and-play jailbreaking detector designed to identify maliciously perturbed image inputs, utilizing the cross-modal similarity between harmful queries and adversarial images. This simple yet effective cross-modality information detector, CIDER\textit{CIDER}CIDER, is independent of the target VLMs and requires less computation cost. Extensive experimental results demonstrate the effectiveness and efficiency of CIDER\textit{CIDER}CIDER, as well as its transferability to both white-box and black-box VLMs.

View on arXiv
Comments on this paper