Ropedia W0 Brings Human Experience to Cross-Embodiment World Models
Ropedia W0 is a world foundation model for egocentric locomotion and manipulation across human and robot embodiments, using camera poses and projected skeletons as a shared motion interface. Its Xp-Data contains 10,000 hours of human egocentric video and 3,000 hours of robot data, while the paper reports state-of-the-art results on five cross-embodiment benchmarks and real-time, minute-long rollouts after distillation.
The paper connects large-scale human experience to robot world modeling, including cross-embodiment prediction, policy evaluation, model-based planning and real-time VR interaction.
Computer Science > Computer Vision and Pattern Recognition
arXiv:2610.26855 (cs)
[Submitted on 19 Oct 2026]
Title:Ropedia W0: Scaling Loco-Manipulation World Models across Embodiments with Human Experience
Authors:Tianqi Liu, Hanfei Zhao, Qifeng Liang, Hui En Pang, Jiarui Xu, Sayang Mu, Penghao Qiu, YingSheng Zhang, Jun Wang, Ziang Cao, Mingze Sun, Yukang Cao, Fangzhou Hong, Zhaoxi Chen, Ziwei Liu
View a PDF of the paper titled Ropedia W0: Scaling Loco-Manipulation World Models across Embodiments with Human Experience, by Tianqi Liu and 14 other authors
#Ropedia#Ropedia W0#Xp-Data#Xp-Teleop#FastWAM#Cosmos3-Nano#Unitree G1#GR-1