The Transferable Latent Operator (TLO) is proposed, which learns a unified flow representation shared by Eulerian field prediction and Lagrangian particle rollout, and consistently outperforms existing neural operators in both Eulerian field prediction and zero-shot Lagrangian rollout.
Abstract
Lagrangian modeling is vital to fluid dynamics, as it characterizes particle transport and complements the Eulerian representation. However, Lagrangian trajectories are less commonly available than Eulerian fields, while most neural operators are trained and evaluated primarily in the Eulerian representation. This mismatch motivates a new learning problem: can a model trained solely on Eulerian observations generalize zero-shot from Eulerian field prediction to Lagrangian particle rollout, without Lagrangian supervision or task-specific adaptation? To address this problem, we propose the Transferable Latent Operator (TLO), which learns a unified flow representation shared by Eulerian field prediction and Lagrangian particle rollout. TLO decouples latent flow evolution from coordinate-dependent decoding: querying the evolving latent representation at fixed spatial coordinates yields Eulerian fields, whereas querying velocities at particle positions and recursively updating these positions enables Lagrangian rollout. Across five fluid-dynamics benchmarks, TLO consistently outperforms existing neural operators in both Eulerian field prediction and zero-shot Lagrangian rollout, with further gains from limited Lagrangian fine-tuning.
Variational modeling is highlighted as a complementary approach to robust long-horizon neural PDE dynamics by introducing latent Markov dynamics, which combines spatially resolved latent fields, structured Gaussian perturbations, and a neural-operator transition.
Jun-Yi Liao, J. Guilleminot, Vahid Tarokh· 0 citations
Grounded in the insight that temporal position sequences implicitly encode underlying dynamics, HiLNN employs a recurrent encoder to extract a latent context from history that not only reconstructs the unobserved initial velocity but also adaptively modulates the mass matrix, potential energy, and damping coefficients...
Tian-Shuo Zhang, Xianglei Xing, Wen-Zhe Zhai et al.· 1 citation
Convolutional Residual Global Interactions (øurs), which projects particle features to an Eulerian grid, applies residual global updates, and maps them back to particles, and outperforms GNS by 47% on average, supporting the hypothesis that jointly modeling coarse and fine interactions is an effective strategy for neur...
Ethan Ji, Yuan-Zhou Chen, Arush Ramteke et al.· Knowledge Discovery and Data...· 1 citation· ⚡1
The results indicate that physically meaningful molecular trajectories can emerge directly from physics-only supervision, supporting the feasibility of trajectory-unsupervised neural solvers for molecular dynamics.
Petros Triantafyllos, P. Krokidas, C. Rekatsinas· 0 citations
To learn the unknown dynamics in the proximity of a small body directly from tracking data, representing them as a feed-forward neural network embedded in the system Hamiltonian, which is complementary to the known part of the Hamiltonian.
A novel Energy-Structured Latent Latent World Model to structure the ELWM latent state to explicitly carry energy and momentum, ensuring strictly causal transitions via dissipation and control ports, and demonstrates that embedding explicit physical structures into latent spaces intrinsically bridges the gap between pr...
Yapeng Liu, Yuan-Zhao Zhai, Bo Ding et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.