Where Identity Lives: Localized, Retain-Free Identity Unlearning in Multimodal Large Language Models

By Kangwook Ko · Paper · cs.CL

Removing a specific individual's information from multimodal large language models (MLLMs) is often needed after deployment, but existing methods rely on a retain set, which is hardest to obtain at that point, and rebuilding it recreates the privacy exposure that unlearning aims

Cs.cl

View original

HomeResourceLoading…