| title | ODEWorld: A Continuous Predictive Architecture via Physical-Time Flow |
|---|---|
| authors | Dongxiu Liu, Haoyi Niu, Peng Cheng, Yuan Gao, Xirui Kang, Sangli Teng, Koushil Sreenath, Xianyuan Zhan |
| date | 2026-07-30 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2607.27924 |
| upvotes | 14 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf code |
| primer | World models |
| abstract | In the physical world we inhabit, space and time are fundamentally continuous. However, existing machine learning paradigms for world modeling are largely confined to discrete-time prediction, thereby exhibiting significant inefficiency in capturing the dynamics of physical world. We introduce Physical-Time Flow (PT-Flow), a novel approach that learns a continuous latent velocity field operating in physical time. Crucially, the underlying dynamics of sequential data are parameterized by an ordinary differential equation (ODE) embedded in a well-structured representation space. Under this paradigm, the prediction of future can be recast as temporal integration via an ODE solver in the compressed latent space. Building upon PT-Flow, we construct ODEWorld, a continuous-time latent world model that is both efficient and versatile. By extracting time-variant features and enforcing ODE properties on both the dynamical representation space and the latent velocity field, ODEWorld effectively addresses the long-standing representation collapse issue in latent world model literature. This also enables high-quality image reconstruction even after long-horizon prediction. Moreover, its continuous nature allows for arbitrary temporal resolution and even backward prediction, which is impossible for most discrete-time models. Lastly, ODEWorld can provide rich planning-oriented information to facilitate downstream policy learning. Comprehensive experiments demonstrate that ODEWorld successfully reconciles planning-conducive dynamics abstraction with visual realism, excelling in both video generation and robotic control. https://dstate.github.io/odeworld_website/{Project Website}. |
cite
@misc{liu2026odeworld,
title={ODEWorld: A Continuous Predictive Architecture via Physical-Time Flow},
author={Dongxiu Liu and Haoyi Niu and Peng Cheng and Yuan Gao and Xirui Kang and Sangli Teng and Koushil Sreenath and Xianyuan Zhan},
year={2026},
eprint={2607.27924},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2607.27924}
}