Removing Backdoors in Pre-trained Models by Regularized Continual Pre-training

Removing Backdoors in Pre-trained Models by Regularized Continual Pre-training Biru Zhu author Ganqu Cui author Yangyi Chen author Yujia Qin author Lifan Yuan author Chong Fu author Yangdong Deng author Zhiyuan Liu author Maosong Sun author Ming Gu author 2023 text journal article Transactions of the Association for Computational Linguistics continuing MIT Press Cambridge, MA periodical academic journal zhu-etal-2023-removing 10.1162/tacl_a_00622 https://aclanthology.org/2023.tacl-1.91/ 2023 11 1608 1623