ImgTrojan: Jailbreaking Vision-Language Models with ONE Image
Published in NAACL 2025, 2025
Recommended citation: Xijia Tao, Shuai Zhong, Lei Li, Qi Liu, and Lingpeng Kong. "ImgTrojan: Jailbreaking Vision-Language Models with ONE Image." Proceedings of NAACL 2025, pages 7048–7063, 2025.
View the paper in the ACL Anthology or browse the code.