Changhua Meng

Self-evolving LLMs can amplify adversarial threats, making every known attack lineage-persistent and exposing critical vulnerabilities that static defenses can't address.

VISTA achieves a remarkable accuracy increase in GUI grounding tasks by leveraging multiple views, transforming how models learn from visual data.

LLM-powered autonomous agents are alarmingly susceptible to multi-turn, context-aware attacks that bypass standard security measures, nearly doubling the risk trigger rate.