Learning Transferable Dynamics Priors from Action to World Modeling

Ze Huang, Jiahui Zhang, Hairuo Liu, Chenxi Zhang, Ran Cheng, Li Zhang

We study action-conditioned world modeling as a scalable way to learn transferable dynamics priors for robot learning. By pretraining a model to predict how actions drive visual scene evolution, the resulting world model captures reusable interaction dynamics beyond appearance-level video generation. Concretely, we pretrain a multi-view interactive base diffusion world model, A2World, on large-scale robot manipulation data with real action annotations. We validate the learned dynamics priors from two complementary perspectives. First, we adapt A2World into a task- or scene-specialized real-world simulator, A2World-sim, whose long-horizon rollouts support simulator-based policy evaluation and scalable what-if analysis by replacing real-robot rollouts with world model rollouts. Second, starting from the same pretrained weights, we adapt A2World into a video-action joint prediction model, A2World-policy, that predicts actions under visual and instruction conditioning. Experiments across simulation benchmarks and real-robot settings demonstrate that action-conditioned world model pretraining yields transferable dynamics priors that benefit both simulator-centric and policy-centric robot learning.