Rolling-WAM: World Action Models with Rolling Imagination

Yinghua Zhou, Junjie Ye, Yiqi Zhao, Hao Dong, Celina Shiyu Wang, Ruohai Ge, Tingyi Yang, Basile Van Hoorick, Gaurav Sukhatme, Vitor Guizilini, Yue Wang · N/A · 2026

Summary

Abstract Summary

World Action Models (WAMs) couple action generation with future visual prediction for robotic manipulation. However, completing the joint video-action denoising process at each replanning cycle incurs substantial latency, delaying action updates and limiting closed-loop responsiveness. We present Rolling-WAM, a formulation that distributes joint denoising across successive replanning cycles. Our method maintains a sliding window of video-action chunks at staggered noise levels. At each step, a rolling noise schedule fully denoises the imminent action chunk for execution, while partially refining farther-future chunks. As the window advances with new camera observations, the retained future chunks continue their denoising process. This distributes the computational cost over time while carrying an evolving visual-action context across chunk boundaries. Evaluations on LIBERO, RoboTwin, and a real-world Unitree G1 humanoid show that Rolling-WAM achieves competitive manipulation performance. By removing the need to denoise the entire prediction horizon from scratch, it delivers a 4.5x steady-state replanning speedup over standard joint WAMs.

Key Points

World Action Models (WAMs) couple action generation with future visual prediction for robotic manipulation.
However, completing the joint video-action denoising process at each replanning cycle incurs substantial latency, delaying action updates and limiting closed-loop responsiveness.
We present Rolling-WAM, a formulation that distributes joint denoising across successive replanning cycles.
Our method maintains a sliding window of video-action chunks at staggered noise levels
At each step, a rolling noise schedule fully denoises the imminent action chunk for execution, while partially refining farther-future chunks.

Method

This section presents Rolling-WAM, a joint video-action model for low-latency closed-loop robotic control. We first formulate the visual-action modeling problem and identify the computational bottleneck of standard joint denoising (Sec. III-A ). We then introduce a rolling denoising mechanism that maintains a sliding window of predictions at staggered noise levels (Sec. III-B ).

Next, we describe the sampling and execution process that distributes computation across control cycles (Sec. III-C ).

Method — Rolling-WAM: World Action Models with Rolling Imagination — Fig. 3: Imagined and observed rollouts. Rolling-WAM’s imagined video (top) and ground-truth observations (bottom) during policy execution on Put Object Cabinet . Columns show successive moments.

Experiments

Our experiments verify whether rolling denoising can accelerate joint video-action prediction while preserving manipulation performance. We compare task success with a broad set of vision-language-action models and world action models on LIBERO, RoboTwin 2.0, and real-world Unitree G1 tasks. Controlled latency measurements quantify the computational savings of rolling denoising.

Ablations examine the roles of window size, training noise schedules, and cross-chunk action attention. We evaluate Rolling-WAM on the following settings.

Experiments — Rolling-WAM: World Action Models with Rolling Imagination — Fig. 6: Real-world humanoid manipulation tasks. Doll Placement, Plate Stacking, and Bead Pouring (top to bottom), illustrated with qualitative examples from Rolling-WAM rollouts. Each row shows task progression from left to right. Insets show egocentric observations.

Sources and Demonstrations

Figures are reproduced from Zhou et al. See the full paper for experimental details and the project page, when the paper links one, for demonstrations and videos.