Open Source
ODEWorld: A Continuous Predictive Architecture via Physical-Time Flow
arXiv cs.RO0 views
arXiv:2607.27924v4 Announce Type: replace-cross 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. Project page: https://dstate.github.io/odeworld_website/.
Source:https://arxiv.org/abs/2607.27924PreviousObserver-Assisted Relative-Velocity Compensation with LPV-$H_\infty$ Robust Correction for 3D Trajectory Tracking of Underactuated Non-Minimum-Phase AUVs under Ocean CurrentsNextParameter-Dependent LMI Synthesis for Semi-Global Differential ISS Trajectory Tracking of Nonholonomic Mobile Robots Under Multiplicative Wheel Slip
Related posts
GetDevEd - open source system design platform
0dev.to Beginners
Latent Projection for Latent Planning
0arXiv cs.RO
ActionUNet: Improving Robustness of VLA Models with Efficient Multi-scale Fine-tuning
0arXiv cs.RO
Spatial Calibration of Diffuse LiDARs
0arXiv cs.RO
Show HN: Lazycow Studio – turns footage you never edited into reels, locally
0Hacker News New
Show HN: Karaoke Is All You Need
0Hacker News Show