CORTEXA
← Browse
arxivcs.ROcs.CV2026-06-25

World Action Models Enable Continual Imitation Learning with Recurrent Generative Replays

Manish Kumar Govind, Dominick Reilly, Smit Patel, Hieu Le, Srijan Das

Going beyond predicting robot actions, World Action Models (WAMs) can also generate future visual observations. We build on this generative capability to propose Recurrent Generative Replay (REGEN), a continual imitation learning framework that synthesizes pseudo-replay trajectories, enabling a robot policy to rehearse previously learned tasks without storing their original human demonstrations. During continual adaptation, REGEN recursively queries the WAM to synthesize pseudo-replay trajectories conditioned only on prior task instructions and current-task observations. Experiments in both simulation and real-world manipulation settings show that REGEN reduces catastrophic forgetting by up to $50\%$ relative to sequential fine-tuning, while approaching the performance of privileged experience replay methods that require access to real replay data. Finally, we analyze the factors limiting generated replay, identifying long-horizon visual degradation and action-observation inconsistency as the primary bottlenecks. Our results establish WAMs as a promising foundation for continual robot learning without stored demonstrations.

View free PDFSource page

Related papers

arxivcs.ROcs.CV2026-06-30

3D Point World Models: Point Completion Enables More Accurate Dynamics Learning

Skand Peri, Hung Nguyen, Chanho Kim, Li Fuxin, Stefan Lee

Learning predictive models of the world enables robotic control through planning, potentially allowing robots to improvise solutions on new tasks. However, large video-based dynamics models lack explicit 3D spatial structure and suffer from geometrically inconsistent long-term ro…

View free PDFSource page
arxivcs.ROcs.CV2026-06-26

DIM-WAM: World-Action Modeling with Diverse Historical Event Memory

Kai Wang, Zhaopeng Gu, Yixiang Chen, Yuan Xu, Qisen Ma, Jiabing Yang, et al.

World-action models have shown promising robot-manipulation performance by jointly predicting future visual states and actions. However, existing methods mainly rely on short-term history and short-horizon future prediction, which is insufficient for long-horizon tasks whose corr…

View free PDFSource page
arxivcs.CVcs.RO2026-07-10

Causally Debiased Latent Action Model for Embodied Action Conditioned World Models

Yufan Wei, Kun Zhou, Lingjun Mao, Zijun Zhang, Ziming Xu, Ziqiao Xi, et al.

Action-conditioned world models (ACWMs) aim to simulate future observations conditioned on embodied actions, offering a promising foundation for robot planning, policy evaluation, and data augmentation. However, learning controllable ACWMs requires large-scale action-labeled data…

View free PDFSource page
arxivcs.ROcs.CV2026-07-14

FlowWAM: Optical Flow as a Unified Action Representation for World Action Models

Yixiang Chen, Peiyan Li, Yuan Xu, Qisen Ma, Jiabing Yang, Kai Wang, et al.

World Action Models (WAMs) are able to leverage pretrained video generators for both world modeling and action prediction. However, directly leveraging such video generators for control raises a new challenge: how to represent actions in a suitable form that aligns with pretraine…

View free PDFSource page
arxivcs.CVcs.RO2026-07-01

ABot-M0.5: Unified Mobility-and-Manipulation World Action Model

Ronghan Chen, Yandan Yang, Zuojin Tang, Dongjie Huo, Tong Lin, Haoning Wu, et al.

Mobile manipulation is a key capability for general-purpose robots, yet remains challenging for current embodied learning methods. VLA policies are typically reactive and lack explicit world modeling, while existing World Action Models (WAMs) are still poorly aligned with the str…

View free PDFSource page