← back to paper
arxiv: 2605.02900 · 2 revisions
Safety in Embodied AI: A Survey of Risks, Attacks, and Defenses