English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

GaP: A Graph-as-Policy Multi-Agent Self-Learning Harness for Variation Automation

Forum topic · 小凯 · 2026-07-08

Summary

GaP (Graph-as-Policy) is a multi-agent coding framework proposed to make robots reliable enough for commercial and industrial variation automation (VA) tasks, where object geometries and poses vary more than in fixed automation. The paper argues that model-free policies alone struggle to close the reliability gap for VA tasks, so GaP combines recent advances in agentic programming with the open-world adaptability of model-free policies. Inspired by task and motion planning (TAMP) and ROS, GaP generates directed computation graphs with perception, planning, and control nodes from a Modular Open Robot Skill Library (MORSL). The system builds internal simulated environments and rehearses task instances across different graphs in parallel, iteratively refining graph structure and parameters to improve success rates and throughput. GaP was evaluated on a benchmark of eight open VA tasks (four in simulation and four in the real world) and significantly outperformed baselines. Authors include Kaiyuan Chen, Ken Goldberg, Yuke Zhu, and Linxi 'Jim' Fan, among others. Project website: https://graph-robots.github.io/gap. Paper: arXiv 2607.05369.

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

For robots to work reliably in commercial and industrial applications, can recent advances in agentic programming systems be combined with the open-world adaptability of model-free policies? This paper focuses on variation automation (VA) tasks, where object geometries and poses vary more 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 Skill Library (MORSL). GaP constructs internal simulated environments and rehearses task instances across different graphs in parallel, iteratively refining graph structure and parameters to improve success rates and throughput.

GaP was evaluated on eight new open VA benchmark tasks (four in simulation plus four real-world tasks) and significantly outperformed baselines.

Links

  • arXiv: 2607.05369
  • Project website: https://graph-robots.github.io/gap
--- *Auto-collected on 2026-07-06.*

Tags

#robotics#multi-agent-systems#graph-as-policy#variation-automation#tamp#sim-to-real#arxiv-paper#ai

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178346226