Paper Overview
Field: AI / Robotics Authors: Kaiyuan Chen, Shuangyu Xie, Letian Fu, Justin Yu, William Pacini, Sandeep Bajamahal, Hudson Kim, Jaimyn Drake, Daehwa Kim, Haoru Xue, Jonathan Francis, Christian Juette, Peter Schaldenbrand, Muhammet Yunus Seker, Ruwan Wickramarachchi, Uksang Yoo, Guanzhi Wang, Adithyavairavan Murali, Balakumar Sundaralingam, S. Shankar Sastry, Spencer Huang, Yuke Zhu, Linxi 'Jim' Fan, Ken Goldberg Published: 2026-07-06 arXiv: 2607.05369
Abstract
To enable robots to work reliably in commercial and industrial applications, can recent advances in agent programming systems be combined with the open-world adaptability of model-free policies? This paper focuses on Variation Automation (VA) tasks, where object geometry and pose variation is greater than in fixed automation. Model-free policies often struggle to bridge the reliability gap for VA tasks.
Inspired by Task-and-Motion Planning (TAMP) and ROS, the authors propose Graph-as-Policy (GaP), a multi-agent coding framework that generates directed computation graphs with perception, planning, and control nodes from a Modular Open Robot Skills Library (MORSL). GaP creates internal simulation environments to rehearse task instances across different graphs in parallel, iteratively optimizing graph structure and parameters to improve success rate and throughput.
GaP was evaluated on a benchmark of 8 new open VA tasks (4 in simulation + 4 in the real world) and significantly outperformed baselines.
Project website: https://graph-robots.github.io/gap