This paper systematically surveys current efforts on the evaluation, attack, and defense of MLLMs' safety on images and text.
Feb 1, 2024
