On Large Language Model Continual Unlearning

Chongyang Gao, Lixu Wang, Kaize Ding +3

While large language models have demonstrated impressive performance across various domains and tasks, their security issues have become increasingly severe. Machine unlearning has…