[2603.09030] PlayWorld: Learning Robot World Models from Autonomous Play

Computer Science > Robotics arXiv:2603.09030 (cs) [Submitted on 9 Mar 2026 (v1), last revised 6 Apr 2026 (this version, v3)] Title:PlayWorld: Learning Robot World Models from Autonomous Play Authors:Tenny Yin, Zhiting Mei, Zhonghe Zheng, Miyu Yamane, David Wang, Jade Sceats, Samuel M. Bateman, Lihan Zha, Apurva Badithela, Ola Shorinwa, Anirudha Majumdar View a PDF of the paper titled PlayWorld: Learning Robot World Models from Autonomous Play, by Tenny Yin and 10 other authors View PDF HTML (experimental) Abstract: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 contac...