Xuan Wu

This survey provides a comprehensive, structured review of safety research in Embodied AI, analyzing attacks and defenses across the entire embodied pipeline to guide the development of safe, robust, and reliable real-world agents.

The paper introduces SafeHarness, a novel, lifecycle-integrated security architecture that significantly reduces unsafe behavior and attack success rates in LLM agents by weaving multiple defense layers directly into the agent's execution pipeline.

The paper introduces NICE, a novel, theory-grounded diagnostic benchmark for assessing the social intelligence of LLMs, which reveals that current frontier models consistently struggle with specific facets of communication.

VLA-Pro is a plug-and-play framework that enhances cross-task generalization in Vision-Language-Action models by storing and dynamically retrieving task-specific procedural memories, achieving significant improvements in both simulation and real-world robotic manipulation.