Exploring Visual Vulnerabilities via Multi-Loss Adversarial Search for Jailbreaking Vision-Language Models

被引：0
|
作者：
Hao, Shuyang [1 ]
Hooi, Bryan [2 ]
Liu, Jun [3 ]
Chang, Kai-Wei [4 ]
Huang, Zi [5 ]
Cai, Yujun [5 ]
机构：
[1] Southeast University, China
[2] National University of Singapore, Singapore
[3] Lancaster University, United Kingdom
[4] University of California, Los Angeles, United States
[5] University of Queensland, Australia
来源：
arXiv |
关键词：
Compilation and indexing terms; Copyright 2025 Elsevier Inc;
D O I：
暂无
中图分类号：
学科分类号：
摘要：
Adversarial machine learning - Semantics
引用
收藏
相关论文
共 50 条
  • [1] JailbreakZoo: Survey, Landscapes, and Horizons in Jailbreaking Large Language and Vision-Language Models
    Jin, Haibo
    Hu, Leyang
    Li, Xinnuo
    Zhang, Peiyan
    Chen, Chonghan
    Zhuang, Jun
    Wang, Haohan
    arXiv,
  • [2] Adversarial Prompt Tuning for Vision-Language Models
    Zhang, Jiaming
    Ma, Xingjun
    Wang, Xin
    Qiu, Lingyu
    Wang, Jiaqi
    Jiang, Yu-Gang
    Sang, Jitao
    COMPUTER VISION - ECCV 2024, PT XLV, 2025, 15103 : 56 - 72
  • [3] Boosting adversarial transferability in vision-language models via multimodal feature heterogeneity
    Chen, Long
    Chen, Yuling
    Ouyang, Zhi
    Dou, Hui
    Zhang, Yangwen
    Sang, Haiwei
    SCIENTIFIC REPORTS, 2025, 15 (01):
  • [4] On Evaluating Adversarial Robustness of Large Vision-Language Models
    Zhao, Yunqing
    Pang, Tianyu
    Du, Chao
    Yang, Xiao
    Li, Chongxuan
    Cheung, Ngai-Man
    Lin, Min
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 36 (NEURIPS 2023), 2023,
  • [5] Exploring Vision-Language Models for Imbalanced Learning
    Wang Y.
    Yu Z.
    Wang J.
    Heng Q.
    Chen H.
    Ye W.
    Xie R.
    Xie X.
    Zhang S.
    International Journal of Computer Vision, 2024, 132 (01) : 224 - 237
  • [6] Efficient Generation of Targeted and Transferable Adversarial Examples for Vision-Language Models via Diffusion Models
    Guo, Qi
    Pang, Shanmin
    Jia, Xiaojun
    Liu, Yang
    Guo, Qing
    IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY, 2025, 20 : 1333 - 1348
  • [7] MixPrompt: Enhancing Generalizability and Adversarial Robustness for Vision-Language Models via Prompt Fusion
    Fan, Hao
    Ma, Zhaoyang
    Li, Yong
    Tian, Rui
    Chen, Yunli
    Gao, Chenlong
    ADVANCED INTELLIGENT COMPUTING TECHNOLOGY AND APPLICATIONS, PT IX, ICIC 2024, 2024, 14870 : 328 - 339
  • [8] Unveiling Vulnerabilities in Large Vision-Language Models: The SAVJ Jailbreak Approach
    Zhang, Gang
    Fan, Xiaowei
    Fang, Jingquan
    Sun, Yanna
    Shi, Xiayang
    Lu, Chunyang
    ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING-ICANN 2024, PT V, 2024, 15020 : 417 - 434
  • [9] VinVL: Revisiting Visual Representations in Vision-Language Models
    Zhang, Pengchuan
    Li, Xiujun
    Hu, Xiaowei
    Yang, Jianwei
    Zhang, Lei
    Wang, Lijuan
    Choi, Yejin
    Gao, Jianfeng
    2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021, 2021, : 5575 - 5584
  • [10] BRAVE: Broadening the Visual Encoding of Vision-Language Models
    Kar, Oguzhan Fatih
    Tonioni, Alessio
    Poklukar, Petra
    Kulshrestha, Achin
    Zamir, Amir
    Tombari, Federico
    COMPUTER VISION - ECCV 2024, PT XVI, 2025, 15074 : 113 - 132
12345