Jie Li

Safety alignment is a fundamental prerequisite for building trustworthy artificial general intelligence. Despite substantial progress in safety alignment techniques, empirical evidence shows that aligned large language models can still produce unsafe responses under minor internal perturbations,...

Evaluating semantic drift is essential for understanding dynamical discourse evolution and opinion formation in online discussions. However, sparse and uneven distributions of event-specific keywords prevent traditional models from capturing post-level semantic drift. Thus, to address this issue,...