Newest papers

AgenticRoboRL: Bringing Coding-Agent Post-Training to Agentic Robotics

Xinxin Liu, Bingxin Xu, Wenrui Bao, Ming Li, Wentao Bao, Chen Chen, Yuzhang Shang · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.28589 · cited by 0 works (OpenAlex) · open-access PDF

Spanish

Los sistemas de robótica de verificación de agentes combinan los planificadores con arnés robóticos para realizar tareas complejas de manipulación del marco. El trabajo reciente mejora estos sistemas a través de mejores arnés y memoria externa, pero se basa en poderosos modelos de fuente cerrada como planificadores y no mejora las propias capacidades de los planificadores. En contraste, los agentes de codificación han mejorado sustancialmente al realizar tareas de post-entrenamiento del modelo en sí. Por lo tanto, investigamos si el post-entrenamiento también puede mejorar a los planificadores robóticos. Sin embargo, aplicarlo a través de los arnés robóticos requiere acomodar diferentes mecanismos de ejecución, incluyendo la invocación de habilidades de VLA y la generación de código de control. Para este fin, introducimos Agen

Machine translation (Meta M2M100); the English original follows.

Abstract

Agentic robotics systems combine planners with robot harnesses to perform complex manipulation tasks. Recent work improves these systems through better harnesses and external memory, but relies on powerful closed-source models as planners and does not improve the planners' own capabilities. In contrast, coding agents have improved substantially by post-training the model itself. We therefore investigate whether post-training can also improve robot planners. However, applying it across robot harnesses requires accommodating different execution mechanisms, including VLA skill invocation and control-code generation. To this end, we introduce AgenticRoboRL, a framework for post-training robot planners across different harnesses. The training recipe combines supervised fine-tuning with multi-turn reinforcement learning. To accommodate different execution mechanisms, harness-specific adapters connect recorded planner interactions with task outcomes, enabling a common RL training pipeline while keeping execution components fixed. On LIBERO-PRO, AgenticRoboRL raises the success rate of a 9B-parameter planner from 18.1% to 71.0% without external memory, approaching the 72.1% achieved by Codex (GPT-5.5) with external memory. These results demonstrate post-training as an effective approach for improving robot planners. We will release the framework and trained planner checkpoints.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Robot, Robotics, Planner, Artificial intelligence, Computer science, Task (project management)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange