PlayWorld: Learning Robot World Models from Autonomous Play

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Yin, Tenny, Mei, Zhiting, Zheng, Zhonghe, Yamane, Miyu, Wang, David, Sceats, Jade, Bateman, Samuel M., Zha, Lihan, Badithela, Apurva, Shorinwa, Ola, Majumdar, Anirudha
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913006543175680
author Yin, Tenny
Mei, Zhiting
Zheng, Zhonghe
Yamane, Miyu
Wang, David
Sceats, Jade
Bateman, Samuel M.
Zha, Lihan
Badithela, Apurva
Shorinwa, Ola
Majumdar, Anirudha
author_facet Yin, Tenny
Mei, Zhiting
Zheng, Zhonghe
Yamane, Miyu
Wang, David
Sceats, Jade
Bateman, Samuel M.
Zha, Lihan
Badithela, Apurva
Shorinwa, Ola
Majumdar, Anirudha
contents Action-conditioned video models offer a promising path to building general-purpose robot simulators that can improve directly from data. Yet, despite training on large-scale robot datasets, current state-of-the-art video models still struggle to predict physically consistent robot-object interactions that are crucial in robotic manipulation. To close this gap, we present PlayWorld, a simple, scalable, and fully autonomous pipeline for training high-fidelity video world simulators from interaction experience. In contrast to prior approaches that rely on success-biased human demonstrations, PlayWorld is the first system capable of learning entirely from unsupervised robot self-play, enabling naturally scalable data collection while capturing complex, long-tailed physical interactions essential for modeling realistic object dynamics. Experiments across diverse manipulation tasks show that PlayWorld generates high-quality, physically consistent predictions for contact-rich interactions that are not captured by world models trained on human-collected data. We further demonstrate the versatility of PlayWorld in enabling fine-grained failure prediction and policy evaluation, with up to 40% improvements over human-collected data. Finally, we demonstrate how PlayWorld enables reinforcement learning in the world model, improving policy performance by 65% in success rates when deployed in the real world.
format Preprint
id arxiv_https___arxiv_org_abs_2603_09030
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle PlayWorld: Learning Robot World Models from Autonomous Play
Yin, Tenny
Mei, Zhiting
Zheng, Zhonghe
Yamane, Miyu
Wang, David
Sceats, Jade
Bateman, Samuel M.
Zha, Lihan
Badithela, Apurva
Shorinwa, Ola
Majumdar, Anirudha
Robotics
Artificial Intelligence
Action-conditioned video models offer a promising path to building general-purpose robot simulators that can improve directly from data. Yet, despite training on large-scale robot datasets, current state-of-the-art video models still struggle to predict physically consistent robot-object interactions that are crucial in robotic manipulation. To close this gap, we present PlayWorld, a simple, scalable, and fully autonomous pipeline for training high-fidelity video world simulators from interaction experience. In contrast to prior approaches that rely on success-biased human demonstrations, PlayWorld is the first system capable of learning entirely from unsupervised robot self-play, enabling naturally scalable data collection while capturing complex, long-tailed physical interactions essential for modeling realistic object dynamics. Experiments across diverse manipulation tasks show that PlayWorld generates high-quality, physically consistent predictions for contact-rich interactions that are not captured by world models trained on human-collected data. We further demonstrate the versatility of PlayWorld in enabling fine-grained failure prediction and policy evaluation, with up to 40% improvements over human-collected data. Finally, we demonstrate how PlayWorld enables reinforcement learning in the world model, improving policy performance by 65% in success rates when deployed in the real world.
title PlayWorld: Learning Robot World Models from Autonomous Play
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2603.09030