Achieving the broad task generalization of embodied models in physical environments remains a central challenge in robot learning. We study this challenge in World Action Models…
LLM-based robotic agent systems can generate programs for robot control, yet their generalization to new tasks often depends on elaborate scaffolding. Expanding their skill repe…
Conversational assistants are expected to help people with diverse backgrounds, preferences, and communication styles. Yet evaluations centered on a single cooperative user prov…
The emergence of writable, cross-session persistent memory in LLM agents introduces a qualitatively different threat landscape from conventional input-centric security concerns,…