The paper demonstrates that learned simulators can fail in two distinct ways when conditions change: long‑horizon drift due to accumulated errors and incorrect responses to interventions on physical parameters. By adding a symplectic integrator to preserve conservative dynamics, rollouts remain stable for up to 100× the training horizon, while encoding physical coupling via explicit linear factorization allows the model to generalize to unseen signs of that coupling. The study shows that stability and counterfactual generalization arise from separate structural choices, enabling designers to impose each property independently.
By Yufeng Wang, Parivesh Priye, Lu Wei, Haibin Ling
arXiv:2606. 13092v3 Announce Type: replace Abstract: Scale buys interpolation; structure buys certifiable transfer.
By Hongbo Wang
arXiv:2608.23526v1 Announce Type: new
Abstract: World models can predict video without learning dynamics that they reliably preserve. We test whether a frozen DreamerV3 trained only on pendulum video...
By Richard Bao
arXiv:2609.06546v1 Announce Type: cross
Abstract: Although machine learning can be used to predict the evolution of physical systems from data, a formulation that learns only the system state at each...
By Sai Siddharth, Maniarasu Ravi
The paper introduces a world model that learns to predict the evolution of physical systems while respecting key physical principles. By hard‑coding a general structure—generating dynamics from the gradient of a learned energy via a fixed reversible operator and imposing constraints on energy, dissipation, and interventions—the model achieves second‑law compatible dissipation, accurate responses to parameter changes, long‑term stability, and robustness to disturbances. Experiments on an electromagnetic cavity, a particle‑in‑cell grid, and shallow‑water fluid demonstrate that the model can recover accurate constitutive functions, distinguish conserving from dissipating regimes, and transfer learned physics to unseen conditions, outperforming unconstrained models.
By Yufeng Wang, Parivesh Priye, Lu Wei, Haibin Ling
arXiv:2607. 03198v1 Announce Type: new Abstract: World models -- compressed latent representations of an environment that support action-conditioned prediction and planning -- are typically presented as a product of modern self-supervised learning.
By Rajat Ghosh
Object-centric world models forecast future videos by evolving a set of entity slots, but the variables receiving dynamics supervision are often unconstrained visual features. We introduce \method{},...
arXiv:2606. 01520v1 Announce Type: new Abstract: A single action-conditioned latent predictive architecture can in principle be trained on the structured state of a driving scene, a robot workspace, or a financial order book.
By Shayan Shokri
arXiv:2608. 10172v1 Announce Type: new Abstract: Mechanistic interpretability explains models by identifying circuits inside them, but has no way to tell whether a circuit is a property of the model or an artifact of the method that found it.
By Ashim Dhor, Pin-Yu Chen
arXiv:2606. 24946v1 Announce Type: new Abstract: Learned world models are useful only over horizons on which their rollout error remains controlled.
By Hongbo Wang
arXiv:2607. 27017v1 Announce Type: new Abstract: A central premise of latent world models is that predicting the future forces a representation to internalize the physics of its environment.
By Kaizhen Tan (New York University, Carnegie Mellon University), Xin Xu (Carnegie Mellon University), Siru Tao (Carnegie Mellon University), Hanzhe Hong (Carnegie Mellon University), Yang Feng (Columbia University), Heqing Du (Columbia University)
The paper proposes storing finite collections of knowledge as the orbit of a single compact latent generator that cycles exactly back to its starting point. By encoding each item as a phase of a fixed rotation in a learned latent space and decoding all phases with a shared network, the method guarantees exact closure through a discrete Fourier operator. Experiments on images and video clips show that this exactly periodic operator outperforms general learned or norm‑preserving operators, achieving comparable or better fidelity while eliminating visible seams and enabling efficient compression.
By Siddharth Pal, Viktoria Rojkova