JEPA4R brings JEPA planning to 4D point clouds and real robots
A newly submitted paper introduces JEPA4R, an embodied JEPA world model that observes single-depth-camera point clouds, predicts and plans in latent space, and acts through a policy head using frozen representations. The authors report a mean point error within 1.1 cm over 200 predicted steps, strong results across OGBench, CortexBench, MetaWorld, and Adroit, and zero-shot operation of a real robotic arm after digital-twin training. These benchmark and hardware results are author-reported and have not been independently confirmed.
JEPA4R connects joint-embedding world modeling with metric 3D observations and latent-space robot planning, while the reported zero-shot transfer from a digital twin to a physical arm offers a concrete embodied-AI use case.
Computer Science > Robotics
arXiv:2611.19771 (cs)
[Submitted on 15 Nov 2026]
Title:JEPA4R: Embodied JEPA World Models with 4D Point Planning
Authors:Jiarui Dong, Zhiheng Liang, Yanjun Li, Zeyu Zhang, Heejeong Nam, Haoyu Chen, Hao Tang
View a PDF of the paper titled JEPA4R: Embodied JEPA World Models with 4D Point Planning, by Jiarui Dong and 6 other authors