When Hallucinated Concepts Cross Modals: Unveiling Backdoor Vulnerability in Multi-modal In-context Learning

Open Webpage

Guanyu Hou, Jiaming He, Yitong Qiao, Jiachen Li, Qiyang Song, Ji Guo, Zihan Wang, Wenbo Jiang

Published: 09 Dec 2025, Last Modified: 11 Mar 2026CrossrefEveryoneRevisionsCC BY-SA 4.0