Machine Unlearning in MLLMs

Machine Unlearning in Vision-Language Models (VLMs) presents unique challenges due to the multi-modal nature of vision and language alignment; below is a brief overview of key considerations.

Notes: Link to the pdf