Affordance-Graphed Task Worlds: Self-Evolving Task Generation for Scalable Embodied Learning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Liu, Xiang, Cui, Sen, Yao, Guocai, Cao, Zhong, Ma, Jingheng, Zhang, Min, Zhang, Changshui
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917270793486336
author Liu, Xiang
Cui, Sen
Yao, Guocai
Cao, Zhong
Ma, Jingheng
Zhang, Min
Zhang, Changshui
author_facet Liu, Xiang
Cui, Sen
Yao, Guocai
Cao, Zhong
Ma, Jingheng
Zhang, Min
Zhang, Changshui
contents Training robotic policies directly in the real world is expensive and unscalable. Although generative simulation enables large-scale data synthesis, current approaches often fail to generate logically coherent long-horizon tasks and struggle with dynamic physical uncertainties due to open-loop execution. To address these challenges, we propose Affordance-Graphed Task Worlds (AGT-World), a unified framework that autonomously constructs interactive simulated environments and corresponding robot task policies based on real-world observations. Unlike methods relying on random proposals or static replication, AGT-World formalizes the task space as a structured graph, enabling the precise, hierarchical decomposition of complex goals into theoretically grounded atomic primitives. Furthermore, we introduce a Self-Evolution mechanism with hybrid feedback to autonomously refine policies, combining Vision-Language Model reasoning and geometric verification. Extensive experiments demonstrate that our method significantly outperforms in success rates and generalization, achieving a self-improving cycle of proposal, execution, and correction for scalable robot learning.
format Preprint
id arxiv_https___arxiv_org_abs_2602_12065
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Affordance-Graphed Task Worlds: Self-Evolving Task Generation for Scalable Embodied Learning
Liu, Xiang
Cui, Sen
Yao, Guocai
Cao, Zhong
Ma, Jingheng
Zhang, Min
Zhang, Changshui
Robotics
Training robotic policies directly in the real world is expensive and unscalable. Although generative simulation enables large-scale data synthesis, current approaches often fail to generate logically coherent long-horizon tasks and struggle with dynamic physical uncertainties due to open-loop execution. To address these challenges, we propose Affordance-Graphed Task Worlds (AGT-World), a unified framework that autonomously constructs interactive simulated environments and corresponding robot task policies based on real-world observations. Unlike methods relying on random proposals or static replication, AGT-World formalizes the task space as a structured graph, enabling the precise, hierarchical decomposition of complex goals into theoretically grounded atomic primitives. Furthermore, we introduce a Self-Evolution mechanism with hybrid feedback to autonomously refine policies, combining Vision-Language Model reasoning and geometric verification. Extensive experiments demonstrate that our method significantly outperforms in success rates and generalization, achieving a self-improving cycle of proposal, execution, and correction for scalable robot learning.
title Affordance-Graphed Task Worlds: Self-Evolving Task Generation for Scalable Embodied Learning
topic Robotics
url https://arxiv.org/abs/2602.12065