Preprint
Oct 2026
RoboWorM: Evaluating Movement Realism, Consistency and Acceleration Affects in Embodied Video World Models
Video world models (VWMs) are rapidly emerging as scalable neural simulators for robotic manipulation. However, most benchmarks only evaluate them as open-world video generators, using Vision-Language Models (VLMs) that lack strict 3D spatial and kinematic reasoning. To bridge this gap, we introduce RoboWorM, a compreh...
Doanh Le Thien, Viet-Thanh Nguyen, Quang Tri Pham et al.
· 0 citations