CarryOn: A Dual-Loop Harness for Personalized and Reliable Robot Assistance
From the abstract
While generalist vision-language-action (VLA) models have enabled diverse open-world manipulation, real-world robotic assistance requires two critical capabilities they currently lack: personalization and reliable execution. First, executing an instruction like "make me a sandwich" requires grounding subtask decomposition in persistent user preferences rather than relying on a single generic solution.
From the abstract. Our summary is in progress.