Dangerous Language Habits! Exploiting Code-Mixing for Backdoor Attacks on NLP Models.

@inproceedings{jin-2025-dangerous, author = {Haotian Jin and Haihui Fan and Jinchao Zhang and Yang Li and Bo Li and Junhao Zhou}, editor = {Meeyoung Cha and Carl Yang and Senjuti Basu Roy and Chanyoung Park and Zhenhui Jessie Li and Noseong Park and Carl Yang and Senjuti Basu Roy and Zhenhui Jessie Li and Jaap Kamps and Kijung Shin and Bryan Hooi and Lifang He}, title = {{Dangerous Language Habits! Exploiting Code-Mixing for Backdoor Attacks on NLP Models}}, booktitle = {{Proceedings of the 34th ACM International Conference on Information and Knowledge Management, CIKM 2025, Seoul, Republic of Korea, November 10-14, 2025}}, pages = {1220--1230}, publisher = {ACM}, year = {2025}, url = {https://doi.org/10.1145/3746252.3761104}, doi = {https://doi.org/10.1145/3746252.3761104} }