Training-free Lexical Backdoor Attacks on Language Models.

Yujin Huang, Terry Yue Zhuo, Qiongkai Xu, Han Hu, Xingliang Yuan, Chunyang Chen

Browse the full WWW paper archive.