MEGA Hub

Can World and Action Modeling Benefit Each Other under Unified Causal Training?

Authors

Do you know Yue Su?You can claim authorship or link another user.Do you know Bokai Lin?You can claim authorship or link another user.Do you know Xinkai Wang?You can claim authorship or link another user.Do you know Chubin Zhang?You can claim authorship or link another user.Do you know Xiangnan Wu?You can claim authorship or link another user.Do you know Kuan Zhang?You can claim authorship or link another user.Do you know Weiyang Jin?You can claim authorship or link another user.Do you know Jiwen Yu?You can claim authorship or link another user.Do you know Yiming Li?You can claim authorship or link another user.Do you know Cewu Lu?You can claim authorship or link another user.Do you know Lixin Yang?You can claim authorship or link another user.Do you know Xihui Liu?You can claim authorship or link another user.

Abstract

World and action modeling are closely related, yet their reciprocal benefits under unified causal training remain unclear. We introduce OneWM, a simple framework for unified causal training that treats actions as both conditions for world prediction and targets for policy learning within a shared video-action architecture. Using this framework, we investigate whether learning forward dynamics conditioned on specified actions improves action learning, and whether learning to generate actions in turn improves action-conditioned world modeling. We use LIBERO, LIBERO-Plus, and RoboCasa365 as embodied testbeds to evaluate both policy learning and world modeling, complemented by real-world policy experiments. Learning the visual consequences of specified actions improves policy performance and generalization across changing scenes. In turn, action learning improves multi-view world-prediction fidelity under the same architecture. Together, these results answer both questions positively, supporting reciprocal benefits between world modeling and action generation under unified causal training.

Community

00