As Multimodal Large Language Models (MLLMs) develop, their potential security issues have become increasingly prominent. Machine Unlearning (MU), as an effective strategy for forgetting specific ...

Abstract: Traditional data augmentation methods are typically computation-driven, randomly selecting words for modification with equal probability for each word. However, these methods do not take ...

Abstract: In recent times, both social media platforms and researchers have been working to identify hateful language using large language models. However, none of these efforts have focused on using ...