mlsanityData to 14 Oct 2026

EddyFormer: Accelerated Neural Simulations of Three-Dimensional Turbulence at Scale

arXiv 2510.24173 ↗First posted 28 Oct 2025cs.LG cs.NA math.DScited by 6

Not on the map. Why?

Abstract (arXiv)

Computationally resolving turbulence remains a central challenge in fluid dynamics due to its multi-scale interactions. Fully resolving large-scale turbulence through direct numerical simulation (DNS) is computationally prohibitive, motivating data-driven machine learning alternatives. In this work, we propose EddyFormer, a Transformer-based spectral-element (SEM) architecture for large-scale turbulence simulation that combines the accuracy of spectral methods with the scalability of the attention mechanism. We introduce an SEM tokenization that decomposes the flow into grid-scale and subgrid-scale components, enabling capture of both local and global features. We create a new three-dimensional isotropic turbulence dataset and train EddyFormer to achieves DNS-level accuracy at 256^3 resolution, providing a 30x speedup over DNS. When applied to unseen domains up to 4x larger than in training, EddyFormer preserves accuracy on physics-invariant metrics-energy spectra, correlation functions, and structure functions-showing domain generalization. On The Well benchmark suite of diverse turbulent flows, EddyFormer resolves cases where prior ML models fail to converge, accurately reproducing complex dynamics across a wide range of physical conditions.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 39 of its 77 references that are in the map. First posted on arXiv 28 Oct 2025.

Builds on

39 of 77 references.

References · 39

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  4. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  5. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,703
  6. 2020Fourier Neural Operator for Parametric Partial Differential Equationscited by 2,908
  7. 2021Characterizing possible failure modes in physics-informed neural networkscited by 643
  8. 2021Machine learning accelerated computational fluid dynamicscited by 471
  9. 2022Fourier Neural Operator with Learned Deformations for PDEs on General Geometriescited by 427
  10. 2021Choose a Transformer: Fourier or Galerkincited by 343
  11. 2023GNOT: A General Neural Operator Transformer for Operator Learningcited by 336
  12. 2024Transolver: A Fast Transformer Solver for PDEs on General Geometriescited by 326
27 more
  1. 2023Geometry-Informed Neural Operator for Large-Scale 3D PDEscited by 303
  2. 2021U-FNO -- An enhanced Fourier neural operator-based deep-learning model for multiphase flowcited by 297
  3. 2022Transformer for Partial Differential Equations' Operator Learningcited by 283
  4. 2021Factorized Fourier Neural Operatorscited by 249
  5. 2019Towards Physics-informed Deep Learning for Turbulent Flow Predictioncited by 204
  6. 2024Poseidon: Efficient Foundation Models for PDEscited by 200
  7. 2022Respecting causality is all you need for training physics-informed neural networkscited by 156
  8. 2024The Well: a Large-Scale Collection of Diverse Physics Simulations for Machine Learningcited by 145
  9. 2023Scalable Transformer for PDE Surrogate Modelingcited by 128
  10. 2023Multiple Physics Pretraining for Physical Surrogate Modelscited by 98
  11. 2025Transolver++: An Accurate Neural Solver for PDEs on Million-Scale Geometriescited by 97
  12. 2024Weak baselines and reporting biases lead to overoptimism in machine learning for fluid-related partial differential equationscited by 91
  13. 2020A fast and accurate physics-informed neural network reduced order model with shallow masked autoencodercited by 80
  14. 2023Multi-Grid Tensorized Fourier Neural Operator for High-Resolution PDEscited by 68
  15. 2024Pretraining Codomain Attention Neural Operators for Solving Multiphysics PDEscited by 59
  16. 2022Learning to correct spectral methods for simulating turbulent flowscited by 49
  17. 2023From Zero to Turbulence: Generative Modeling for 3D Flow Simulationcited by 49
  18. 2023Group Equivariant Fourier Neural Operators for Partial Differential Equationscited by 45
  19. 2022Learned Turbulence Modelling with Differentiable Fluid Solvers: Physics-based Loss-functions and Optimisation Horizonscited by 39
  20. 2021Deep Learning for Reduced Order Modelling and Efficient Temporal Evolution of Fluid Simulationscited by 36
  21. 2018Deep Neural Networks for Data-Driven Turbulence Modelscited by 35
  22. 2022Learning differentiable solvers for systems with hard constraintscited by 34
  23. 2024Scaling physics-informed hard constraints with mixture-of-expertscited by 24
  24. 2023Neural Spectral Methods: Self-supervised learning in the spectral domaincited by 14
  25. 2025Simulating Three-dimensional Turbulence with Physics-informed Neural Networkscited by 14
  26. 2023Neural Ideal Large Eddy Simulation: Modeling Turbulence with Neural Stochastic Differential Equationscited by 13
  27. 2022Spectrally Adapted Physics-Informed Neural Networks for Solving Unbounded Domain Problemscited by 9

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 5.

6 papers.

  1. 2026PDEInvBench: A Comprehensive Dataset and Design Space Exploration of Neural Networks for PDE Inverse Problemscited by 2
  2. 2025CORGI: GNNs with Convolutional Residual Global Interactions for Lagrangian Simulationcited by 1
  3. 2026A Multimodal Vision Transformer-based Modeling Framework for Prediction of Fluid Flows in Energy Systemscited by 1
  4. 2026Breakeven complexity: A new perspective on neural partial differential equation solverscited by 0
  5. 2026Transolver-σ: Joint Spectral-Physical Subspace Modeling for Neural PDE Solvingcited by 0
  6. 2026Scale-Split Neural Operator for Memory- and Data-Efficient 3D Turbulence Predictioncited by 0