LATIDIA · Robótica

ChunkAct: fragmentación de acciones para robótica genética

arXiv:2610.18874v1 Tipo de anuncio: nuevo Resumen: Agentic Robotics coloca un agente de modelo de lenguaje grande (LLM) en el bucle de control del robot, donde compone herramientas de percepción, movimiento y visión-lenguaje-acción (VLA) bajo el

La noticia

arXiv:2610.18874v1 Announce Type: new Abstract: Agentic robotics places a large language model (LLM) agent in the robot control loop, where it composes perception, motion, and vision-language-action (VLA) tools under the guidance of a memory and thereby generalizes to new tasks without retraining. This generality comes at a steep price in efficiency. Every primitive requires a round of LLM reasoning over the full interaction history, so the cost of an episode grows with the number of primitives. We find this reasoning largely redundant: once a sub-goal is chosen, its primitives follow a routine already prescribed by memory. We present ChunkAct, which introduces action chunking to agentic robotics. Online, a single call