analiza aiqbench
This research addresses a critical gap in general-purpose household robotics by introducing MAGE, a data synthesis pipeline that extracts long-horizon planning and grounding capabilities directly from appliance manuals. The resulting UseAppliance dataset enables AppliancePlan to achieve significant performance gains on real robots with minimal parameters, demonstrating effective sim-to-real transfer for complex tasks like washing dishes or making coffee. This work advances the field by providing scalable methods to train models that can robustly manipulate diverse appliances without relying solely on expensive physical data collection.
Scaling Manual-Grounded Appliance Manipulation with Data Synthesis and Unified Planning
arXiv:2608.15863v1 Announce Type: new Abstract: Operating household appliances requires long-horizon planning that is state-dependent and robust to disturbances, yet existing large models fall short, as no sufficiently diverse, task-oriented dataset exists to support such planning. To bridge this gap, we propose MAGE, a scalable data synthesis pipeline that introduces a novel Hierarchical Appliance Graph (HAG) to automatically generate part grounding, long-horizon planning, and closed-loop recovery data from appliance manuals. With MAGE, we build UseAppliance, the first large-scale dataset for manual-grounded appliance manipulation planning, spanning 22 appliance categories with 89K+ part annotations, 53K+ manipulation tasks, and 33K+ closed-loop adjustment steps. Built on UseAppliance, we develop AppliancePlan, an end-to-end model for manual-grounded appliance manipulation planning. On RealAppliance-Bench, AppliancePlan with only 7B parameters achieves over 10x the best baseline on open-loop planning and consistently outperforms state-of-the-art models across all tasks. Real-robot experiments on six household appliances further confirm effective sim-to-real transfer, marking an important step toward general-purpose household robotics.
Abstract:Operating household appliances requires long-horizon planning that is state-dependent and robust to disturbances, yet existing large models fall short, as no sufficiently diverse, task-oriented dataset exists to support such planning. To bridge this gap, we propose MAGE, a scalable data synthesis pipeline that introduces a novel Hierarchical Appliance Graph (HAG) to automatically generate part grounding, long-horizon planning, and closed-loop recovery data from appliance manuals. With MAGE, we build UseAppliance, the first large-scale dataset for manual-grounded appliance manipulation planning, spanning 22 appliance categories with 89K+ part annotations, 53K+ manipulation tasks, and 33K+ closed-loop adjustment steps. Built on UseAppliance, we develop AppliancePlan, an end-to-end model for manual-grounded appliance manipulation planning. On RealAppliance-Bench, AppliancePlan with only 7B parameters achieves over 10x the best baseline on open-loop planning and consistently outperforms state-of-the-art models across all tasks. Real-robot experiments on six household appliances further confirm effective sim-to-real transfer, marking an important step toward general-purpose household robotics.
Robotics (cs.RO); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM)
arXiv:2608.15863 [cs.RO]
(or arXiv:2608.15863v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2608.15863 arXiv-issued DOI via DataCite (pending registration)
Text integral sub licența deschisă a publicației · 183 cuvinte
Subiecte conexe
News
Quantum Computing Report · 25 iul. 2026
Building trust in AI agents: A production-ready evaluation framework with Amazon Bedrock AgentCore
AWS · 16 oct. 2026
News
Quantum Computing Report · 7 iun. 2026
SciHorizon-eLab: An Agentic Protocol-to-Task Compiler for Scalable Benchmarking of Scientific Embodied Agents
arXiv · 28 sept. 2026