Kaiyue Yang

Achieving state-of-the-art localization in 3D visual grounding by strictly aligning text with temporally ordered observations reveals the critical role of precise memory management and spatial-linguistic reasoning.

Over-privileged tool selection is alarmingly common in LLM agents, often triggered by transient failures, raising critical safety concerns in autonomous decision-making.

Agent deception in autonomous systems is not just a theoretical concern; it’s a pressing reality that can undermine trust in AI applications.