pith. sign in

arxiv: 2605.16573 · v1 · pith:XM4XFC2Fnew · submitted 2026-05-15 · 💻 cs.LG · cs.AI· physics.flu-dyn

Wavelet Flow Matching for Multi-Scale Physics Emulation

Gabriele Accarino , Juan Nathaniel , Carla Roesch , Pierre Gentine , Sara Shamekh , Duncan Watson-Parris , Viviana Acquaviva This is my paper

Pith reviewed 2026-05-20 19:28 UTC · model grok-4.3

classification 💻 cs.LG cs.AIphysics.flu-dyn
keywords wavelet flow matchingmulti-scale emulationgenerative modelingfluid dynamicsoptimal transportPDE emulationchaotic systemsU-Net
0
0 comments X

The pith

Wavelet Flow Matching performs optimal transport directly in multi-scale wavelet space to emulate chaotic fluid dynamics with improved stability and detail preservation.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper presents Wavelet Flow Matching as a generative emulator that avoids both the over-smoothing of deterministic models and the extra cost of latent-space autoencoders. It does this by carrying out flow matching on a hierarchical wavelet representation of the data, using a U-Net to predict the transport velocities at each scale. The authors test the approach on three chaotic fluid systems and report better long-horizon stability, accuracy, and preservation of fine-scale spectral properties than current state-of-the-art methods. A reader would care because the method offers a practical route to stable, detail-preserving simulations of multi-scale physical systems without separate pre-training stages.

Core claim

Wavelet Flow Matching enables generative emulation of multi-scale PDE-governed systems by performing optimal transport directly in the hierarchical wavelet domain, using a U-Net to jointly predict transport velocities for a prescribed wavelet representation; this yields superior long-horizon stability, accuracy, and spectral coherence on chaotic fluid dynamics compared with prior models while eliminating the need for a separately trained autoencoder.

What carries the argument

The hierarchical wavelet representation paired with U-Net velocity prediction inside the flow-matching framework, which supplies a training-free multi-scale space for optimal transport.

If this is right

  • Long autoregressive rollouts become feasible for fluid emulators without progressive loss of fine-scale energy.
  • Generative models can operate directly on wavelet coefficients, removing the separate pre-training step for latent compression.
  • Spectral coherence is retained across scales because the wavelet basis already encodes the multi-resolution structure.
  • The same architecture can be applied to other PDE systems that exhibit clear scale separation.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The approach may generalize to non-fluid multi-scale problems such as atmospheric or ocean modeling where wavelet decompositions are already used numerically.
  • Because the wavelet space is fixed rather than learned, it could be combined with existing wavelet-based numerical solvers to create hybrid emulators.
  • Replacing learned latents with an explicit hierarchical basis might reduce training data requirements in other flow-matching applications.
  • Extensions could test whether the same U-Net velocity predictor works when the wavelet family or decomposition depth is varied.

Load-bearing premise

The hierarchical wavelet representation combined with U-Net velocity prediction is sufficient to capture the essential multi-scale transport without requiring a separately trained autoencoder or suffering from information loss at fine scales.

What would settle it

A controlled experiment on a fourth chaotic fluid system in which long-horizon rollout accuracy and power-spectrum fidelity are measured against the same baselines; degradation below current methods would falsify the sufficiency claim.

Figures

Figures reproduced from arXiv: 2605.16573 by Carla Roesch, Duncan Watson-Parris, Gabriele Accarino, Juan Nathaniel, Pierre Gentine, Sara Shamekh, Viviana Acquaviva.

Figure 1
Figure 1. Figure 1: Illustration of the Wavelet Flow Matching architecture. During training, Gaussian noise is [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Qualitative results at different rollout snapshots ( [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: VRMSE (a) and CRPS (b) across rollout steps, and Spectral coherence RMSE (c) across [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
read the original abstract

Accurate emulation of multi-scale physical systems governed by PDEs demands models that remain stable over long autoregressive rollouts while preserving fine-scale structures. Deterministic emulators produce overly-smoothed predictions, while generative approaches better capture details but are costly. Latent-space generative models have emerged as a compromise but with the additional cost of separately pre-trained autoencoders. We propose Wavelet Flow Matching (WFM), a novel generative emulator that overcomes current trade-offs between cost and skill by performing optimal-transport directly in the multi-scale wavelet space. Rather than learning a latent compression, WFM leverages the hierarchical structure of a U-Net to jointly predict transport velocities of a prescribed wavelet representation. On three challenging systems of chaotic fluid dynamics, WFM achieves superior long-horizon stability, accuracy and spectral coherence compared to state-of-the-art models. Our results clearly position the wavelet space as an effective training-free representation for generative emulation of complex physical dynamics.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 2 minor

Summary. The paper proposes Wavelet Flow Matching (WFM), a generative emulator for multi-scale PDE-governed physical systems that performs flow matching directly in an invertible hierarchical wavelet representation. A U-Net is used to predict transport velocities in this space, avoiding the need for a separately trained autoencoder. The central claim is that WFM achieves superior long-horizon stability, accuracy, and spectral coherence compared to state-of-the-art models on three challenging chaotic fluid dynamics systems.

Significance. If the performance claims hold under rigorous validation, the work would be significant for offering a training-free multi-scale representation that balances computational cost and fidelity in generative emulation of chaotic fluids. The approach leverages the invertibility of wavelets to jointly handle scales within a single U-Net, which could reduce overhead relative to latent-space methods while preserving fine-scale structures.

major comments (1)
  1. [§4] §4 (Experimental results): The reported superior performance on the three fluid systems lacks accompanying quantitative tables with error bars, explicit descriptions of baseline implementations (including hyperparameters and training details), and information on data splits or train/test partitioning. These omissions are load-bearing for the central claim of improved long-horizon stability, accuracy, and spectral coherence, as they prevent independent assessment of the gains.
minor comments (2)
  1. [Abstract] The abstract and introduction could more clearly distinguish the proposed method from prior wavelet-based or flow-matching approaches in physics emulation to highlight novelty.
  2. [Methods] Notation for the wavelet decomposition levels and the velocity field prediction could be standardized with explicit definitions early in the methods section to aid readability.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their constructive review and for highlighting areas where the experimental presentation can be strengthened. We address the major comment below and will incorporate the suggested improvements in the revised manuscript.

read point-by-point responses

  1. Referee: [§4] §4 (Experimental results): The reported superior performance on the three fluid systems lacks accompanying quantitative tables with error bars, explicit descriptions of baseline implementations (including hyperparameters and training details), and information on data splits or train/test partitioning. These omissions are load-bearing for the central claim of improved long-horizon stability, accuracy, and spectral coherence, as they prevent independent assessment of the gains.

    Authors: We agree that quantitative tables with error bars and explicit baseline details are necessary to support the central claims. In the revised manuscript we will add a new table in Section 4 that reports mean values and standard deviations (computed over multiple independent random seeds) for all key metrics—long-horizon rollout error, spectral coherence, and stability indicators—across the three fluid systems and all compared methods. We will also add an appendix that provides complete implementation details for every baseline, including exact hyperparameter values, optimizer settings, training schedules, and computational resources. The data partitioning protocol is already stated in Section 3, but we will make this information more prominent by adding a dedicated paragraph that specifies the exact train/test split ratios, any temporal or spatial partitioning, and preprocessing steps. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation self-contained on standard flow matching and wavelets

full rationale

The paper constructs WFM by applying flow-matching transport velocities directly to a fixed, invertible wavelet decomposition of the input fields, with a U-Net predicting those velocities in the hierarchical wavelet basis. This setup is presented as a direct combination of existing optimal-transport flow matching and standard discrete wavelet transforms, without any fitted parameter being renamed as a prediction, without self-definitional equations, and without load-bearing self-citations that close the central argument. Experimental claims of improved long-horizon stability and spectral coherence are obtained from separate rollouts on three external chaotic-fluid benchmarks and do not reduce to the training objective by construction. The derivation therefore remains independent of its reported outputs.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The approach relies on standard wavelet properties and flow-matching theory; no new free parameters, axioms, or invented entities are introduced in the abstract description.

axioms (1)
  • domain assumption Wavelet transforms provide a lossless hierarchical multi-scale decomposition suitable for velocity prediction in chaotic flows.
    Invoked when stating that the method leverages the hierarchical structure without separate autoencoder training.

pith-pipeline@v0.9.0 · 5710 in / 1223 out tokens · 30575 ms · 2026-05-20T19:28:01.528886+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

61 extracted references · 61 canonical work pages · 2 internal anchors

  1. [1]

    Pope.Turbulent Flows

    Stephen B. Pope.Turbulent Flows. Cambridge University Press, 2000

    work page 2000

  2. [2]

    Nature, 619, 533–538, https://doi.org/10.1038/s41586-023-06185-3

    Kaifeng Bi, Lingxi Xie, Hengheng Zhang, Xin Chen, Xiaotao Gu, and Qi Tian. Accurate medium-range global weather forecasting with 3D neural networks.Nature, 619(7970):533– 538, July 2023. ISSN 1476-4687. doi: 10.1038/s41586-023-06185-3. URL https://doi. org/10.1038/s41586-023-06185-3

    work page doi:10.1038/s41586-023-06185-3 2023

  3. [3]

    Chaosbench: A multi-channel, physics-based benchmark for subseasonal- to-seasonal climate prediction.Advances in Neural Information Processing Systems, 37: 43715–43729, 2024

    Juan Nathaniel, Yongquan Qu, Tung Nguyen, Sungduk Yu, Julius Busecke, Aditya Grover, and Pierre Gentine. Chaosbench: A multi-channel, physics-based benchmark for subseasonal- to-seasonal climate prediction.Advances in Neural Information Processing Systems, 37: 43715–43729, 2024

    work page 2024

  4. [4]

    Learning skillful medium-range global weather forecasting.Science, 2023

    Remi Lam et al. Learning skillful medium-range global weather forecasting.Science, 2023

    work page 2023

  5. [5]

    The well: a large-scale collection of diverse physics sim- ulations for machine learning.Advances in Neural Information Processing Systems (NeurIPS), 2024

    Ruben Ohana, Michael McCabe, et al. The well: a large-scale collection of diverse physics sim- ulations for machine learning.Advances in Neural Information Processing Systems (NeurIPS), 2024

    work page 2024

  6. [6]

    Tim N. Palmer. Stochastic weather and climate models.Nature Reviews Physics, 2019

    work page 2019

  7. [7]

    Smith, Ayya Alieva, Qing Wang, Michael P

    Dmitrii Kochkov, Jamie A. Smith, Ayya Alieva, Qing Wang, Michael P. Brenner, and Stephan Hoyer. Machine learning–accelerated computational fluid dynamics.Proceedings of the National Academy of Sciences, 2021

    work page 2021

  8. [8]

    Fourier neural operator for parametric partial differen- tial equations.International Conference on Learning Representations (ICLR), 2021

    Zongyi Li, Nikola Kovachki, Kamyar Azizzadenesheli, Burigede Liu, Kaushik Bhattacharya, Andrew Stuart, and Anima Anandkumar. Fourier neural operator for parametric partial differen- tial equations.International Conference on Learning Representations (ICLR), 2021

    work page 2021

  9. [9]

    Multiple physics pretraining for spa- tiotemporal surrogate models.Advances in Neural Information Processing Systems (NeurIPS), 2024

    Michael McCabe, Bruno Regaldo-Saint Blancard, et al. Multiple physics pretraining for spa- tiotemporal surrogate models.Advances in Neural Information Processing Systems (NeurIPS), 2024

    work page 2024

  10. [10]

    Kossaifi, N

    Jean Kossaifi, Nikola Kovachki, Kamyar Azizzadenesheli, and Anima Anandkumar. Multi-grid tensorized fourier neural operator for high-resolution PDEs.arXiv preprint arXiv:2310.00120, 2023

    work page arXiv 2023

  11. [11]

    Convolutional neural operators for robust and accurate learning of pdes.Advances in Neural Information Processing Systems, 36: 77187–77200, 2023

    Bogdan Raonic, Roberto Molinaro, Tim De Ryck, Tobias Rohner, Francesca Bartolucci, Rima Alaifari, Siddhartha Mishra, and Emmanuel De Bézenac. Convolutional neural operators for robust and accurate learning of pdes.Advances in Neural Information Processing Systems, 36: 77187–77200, 2023

    work page 2023

  12. [12]

    U-Net: Convolutional networks for biomedical image segmentation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-Net: Convolutional networks for biomedical image segmentation. InMICCAI, 2015

    work page 2015

  13. [13]

    Transformer for partial differential equations’ operator learning

    Zijie Li, Kazem Meidani, and Amir Barati Farimani. Transformer for partial differential equations’ operator learning. InTMLR, 2023

    work page 2023

  14. [14]

    Towards stability of autoregressive neural operators.Transactions on Machine Learning Research (TMLR), 2023

    Michael McCabe, Peter Harrington, Shashank Subramanian, and Jed Brown. Towards stability of autoregressive neural operators.Transactions on Machine Learning Research (TMLR), 2023

    work page 2023

  15. [15]

    Veeling, Paris Perdikaris, Richard E

    Phillip Lippe, Bastiaan S. Veeling, Paris Perdikaris, Richard E. Turner, and Johannes Brand- stetter. PDE-Refiner: Achieving accurate long rollouts with neural pde solvers. InNeurIPS, 2023

    work page 2023

  16. [16]

    Benchmarking autoregressive conditional diffusion models for turbulent flow simulation

    Georg Kohl, Li-Wei Chen, and Nils Thuerey. Benchmarking autoregressive conditional diffusion models for turbulent flow simulation. InICML AI4Science Workshop, 2024

    work page 2024

  17. [17]

    High-resolution image synthesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Björn Ommer. High-resolution image synthesis with latent diffusion models. InCVPR, 2022

    work page 2022

  18. [18]

    On conditional diffusion models for PDE simulations

    Aliaksandra Shysheya et al. On conditional diffusion models for PDE simulations. InNeurIPS, 2024. 11

    work page 2024

  19. [19]

    Diffusionpde: Generative pde-solving under partial observation.Advances in Neural Information Processing Systems, 37: 130291–130323, 2024

    Jiahe Huang, Guandao Yang, Zichen Wang, and Jeong Joon Park. Diffusionpde: Generative pde-solving under partial observation.Advances in Neural Information Processing Systems, 37: 130291–130323, 2024

    work page 2024

  20. [20]

    Lost in latent space: An empirical study of latent diffusion models for physics emulation

    François Rozet, Ruben Ohana, Michael McCabe, Gilles Louppe, François Lanusse, and Shirley Ho. Lost in latent space: An empirical study of latent diffusion models for physics emulation. NeurIPS, 2025

    work page 2025

  21. [21]

    Generative emulation of chaotic dynamics with coherent prior.Computer Methods in Applied Mechanics and Engineering, 448:118410, 2026

    Juan Nathaniel and Pierre Gentine. Generative emulation of chaotic dynamics with coherent prior.Computer Methods in Applied Mechanics and Engineering, 448:118410, 2026

    work page 2026

  22. [22]

    Yaron Lipman, Ricky T. Q. Chen, Heli Ben-Hamu, Maximilian Nickel, and Matt Le. Flow matching for generative modeling. InICLR, 2023

    work page 2023

  23. [23]

    Yaron Lipman, Marton Havasi, Peter Holderrieth, Neta Shaul, Matt Le, Brian Karrer, Ricky T. Q. Chen, David Lopez-Paz, Heli Ben-Hamu, and Itai Gat. Flow matching guide and code,

    work page

  24. [24]

    URLhttps://arxiv.org/abs/2412.06264

    work page internal anchor Pith review Pith/arXiv arXiv

  25. [25]

    Flow straight and fast: Learning to generate and transfer data with rectified flow

    Xingchao Liu, Chengyue Gong, and Qiang Liu. Flow straight and fast: Learning to generate and transfer data with rectified flow. InICLR, 2023

    work page 2023

  26. [26]

    Improving and generalizing flow-based generative models with minibatch optimal transport.Transactions on Machine Learning Research (TMLR), 2024

    Alexander Tong, Kilian Fatras, Nikolay Malkin, et al. Improving and generalizing flow-based generative models with minibatch optimal transport.Transactions on Machine Learning Research (TMLR), 2024

    work page 2024

  27. [27]

    Generative latent neural PDE solver using flow matching.arXiv preprint arXiv:2503.22600, 2025

    Zijie Li, Anthony Zhou, and Amir Barati Farimani. Generative latent neural PDE solver using flow matching.arXiv preprint arXiv:2503.22600, 2025

    work page arXiv 2025

  28. [28]

    Efficiency vs

    Srishti Gupta and Yashasvee Taiwade. Efficiency vs. fidelity: A comparative analysis of diffusion probabilistic models and flow matching on low-resource hardware, 2025. URL https://arxiv.org/abs/2511.19379

    work page arXiv 2025

  29. [29]

    Flow matching in latent space

    Quan Dao, Hao Phung, Binh Nguyen, and Anh Tran. Flow matching in latent space.arXiv preprint arXiv:2307.08698, 2023

    work page arXiv 2023

  30. [30]

    From Fourier to neural ODEs: Flow matching for modeling complex systems

    Xin Li, Jingdong Zhang, Qunxi Zhu, Chengli Zhao, Xue Zhang, Xiaojun Duan, and Wei Lin. From Fourier to neural ODEs: Flow matching for modeling complex systems. In Ruslan Salakhutdinov, Zico Kolter, Katherine Heller, Adrian Weller, Nuria Oliver, Jonathan Scarlett, and Felix Berkenkamp, editors,Proceedings of the 41st International Conference on Machine Lea...

    work page 2024

  31. [31]

    Academic Press, 3rd edition, 2009

    Stéphane Mallat.A Wavelet Tour of Signal Processing: The Sparse Way. Academic Press, 3rd edition, 2009

    work page 2009

  32. [32]

    A visual dive into conditional flow matching

    Anne Gagneux, Ségolène Martin, Rémi Emonet, Quentin Bertrand, and Mathurin Massias. A visual dive into conditional flow matching. InICLR Blogposts 2025, 2025. URL https:// iclr-blogposts.github.io/2025/blog/conditional-flow-matching/ . https://iclr- blogposts.github.io/2025/blog/conditional-flow-matching/

    work page 2025

  33. [33]

    Murphy, and Tim Salimans

    Ruiqi Gao, Emiel Hoogeboom, Jonathan Heek, Valentin De Bortoli, Kevin P. Murphy, and Tim Salimans. Diffusion meets flow matching: Two sides of the same coin. 2024. URL https://diffusionflow.github.io/

    work page 2024

  34. [34]

    Stéphane G. Mallat. A theory for multiresolution signal decomposition: the wavelet repre- sentation.IEEE Transactions on Pattern Analysis and Machine Intelligence, 11(7):674–693, 1989

    work page 1989

  35. [35]

    SIAM, Philadelphia, PA, 1992

    Ingrid Daubechies.Ten Lectures on Wavelets. SIAM, Philadelphia, PA, 1992

    work page 1992

  36. [36]

    Adaptive wavelet distillation from neural networks through interpretations

    Wooseok Ha, Chandan Singh, Francois Lanusse, Srigokul Upadhyayula, and Bin Yu. Adaptive wavelet distillation from neural networks through interpretations. InAdvances in Neural Information Processing Systems, volume 34, 2021. 12

    work page 2021

  37. [37]

    Wavesim: A wavelet-based multi-scale similarity metric for weather and climate fields.arXiv preprint arXiv:2512.14656, 2025

    Gabriele Accarino, Viviana Acquaviva, Sara Shamekh, Duncan Watson-Parris, and David Lawrence. Wavesim: A wavelet-based multi-scale similarity metric for weather and climate fields.arXiv preprint arXiv:2512.14656, 2025

    work page arXiv 2025

  38. [38]

    Simoncelli and Edward H

    Eero P. Simoncelli and Edward H. Adelson. Noise removal via bayesian wavelet coring. Proceedings of the IEEE International Conference on Image Processing, 1:379–382, 1996

    work page 1996

  39. [39]

    Donoho and Iain M

    David L. Donoho and Iain M. Johnstone. Ideal spatial adaptation by wavelet shrinkage. Biometrika, 81(3):425–455, 1994

    work page 1994

  40. [40]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. InProceedings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016

    work page 2016

  41. [41]

    Film: visual reasoning with a general conditioning layer

    Ethan Perez, Florian Strub, Harm de Vries, Vincent Dumoulin, and Aaron Courville. Film: visual reasoning with a general conditioning layer. InProceedings of the Thirty-Second AAAI Conference on Artificial Intelligence and Thirtieth Innovative Applications of Artificial Intelligence Conference and Eighth AAAI Symposium on Educational Advances in Artificial...

    work page 2018

  42. [42]

    Multiphase gas and the fractal nature of radiative turbulent mixing layers.The Astrophysical Journal Letters, 894(2):L24, 2020

    Drummond B Fielding, Eve C Ostriker, Greg L Bryan, and Adam S Jermyn. Multiphase gas and the fractal nature of radiative turbulent mixing layers.The Astrophysical Journal Letters, 894(2):L24, 2020

    work page 2020

  43. [43]

    Dedalus: A flexible framework for numerical simulations with spectral methods.Physical Review Research, 2(2):023068, 2020

    Keaton J Burns, Geoffrey M Vasil, Jeffrey S Oishi, Daniel Lecoanet, and Benjamin P Brown. Dedalus: A flexible framework for numerical simulations with spectral methods.Physical Review Research, 2(2):023068, 2020

    work page 2020

  44. [44]

    Learning fast, accurate, and stable closures of a kinetic theory of an active fluid.Journal of Computational Physics, 504:112869, 2024

    Suryanarayana Maddu, Scott Weady, and Michael J Shelley. Learning fast, accurate, and stable closures of a kinetic theory of an active fluid.Journal of Computational Physics, 504:112869, 2024

    work page 2024

  45. [45]

    Fourierflow: Frequency-aware flow matching for generative turbulence modeling, 2025

    Haixin Wang, Jiashu Pan, Hao Wu, Fan Zhang, and Tailin Wu. Fourierflow: Frequency-aware flow matching for generative turbulence modeling, 2025. URL https://arxiv.org/abs/ 2506.00862

    work page arXiv 2025

  46. [46]

    Wavelet diffusion neural operator.arXiv preprint arXiv:2412.04833, 2024

    Peiyan Hu, Rui Wang, Xiang Zheng, Tao Zhang, Haodong Feng, Ruiqi Feng, Long Wei, Yue Wang, Zhi-Ming Ma, and Tailin Wu. Wavelet diffusion neural operator.arXiv preprint arXiv:2412.04833, 2024

    work page arXiv 2024

  47. [47]

    Fourier Neural Operator for Parametric Partial Differential Equations

    Zongyi Li, Nikola Kovachki, Kamyar Azizzadenesheli, Burigede Liu, Kaushik Bhattacharya, Andrew Stuart, and Anima Anandkumar. Fourier neural operator for parametric partial differen- tial equations.arXiv preprint arXiv:2010.08895, 2020

    work page internal anchor Pith review Pith/arXiv arXiv 2010

  48. [48]

    Tapas Tripura and Souvik Chakraborty. Wavelet neural operator for solving parametric partial differential equations in computational mechanics problems.Computer Methods in Applied Mechanics and Engineering, 404:115783, 2023

    work page 2023

  49. [49]

    Tilmann Gneiting and Adrian E. Raftery. Strictly proper scoring rules, prediction, and estimation. Journal of the American Statistical Association, 102(477):359–378, 2007

    work page 2007

  50. [50]

    Shinya Maeyama and Tomo-Hiko Watanabe. Extracting and modeling the effects of small-scale fluctuations on large-scale fluctuations by mori–zwanzig projection operator method.Journal of the Physical Society of Japan, 89(2):024401, 2020

    work page 2020

  51. [51]

    Scientific machine learning for closure models in multiscale problems: A review.arXiv preprint arXiv:2403.02913, 2024

    Benjamin Sanderse, Panos Stinis, Romit Maulik, and Shady E Ahmed. Scientific machine learning for closure models in multiscale problems: A review.arXiv preprint arXiv:2403.02913, 2024

    work page arXiv 2024

  52. [52]

    Deep koopman operators for causal discovery.Communications Physics, 8(1): 513, 2025

    Juan Nathaniel, Carla Roesch, Jatan Buch, Derek DeSantis, Adam Rupe, Kara D Lamb, and Pierre Gentine. Deep koopman operators for causal discovery.Communications Physics, 8(1): 513, 2025. 13

    work page 2025

  53. [53]

    Integrating neural operators with diffusion models improves spectral representation in turbulence modeling.arXiv preprint arXiv:2409.08477, 2024

    Vivek Oommen, Aniruddha Bora, Zhen Zhang, and George Em Karniadakis. Integrating neural operators with diffusion models improves spectral representation in turbulence modeling.arXiv preprint arXiv:2409.08477, 2024

    work page arXiv 2024

  54. [54]

    Spatiotemporal pyramid flow matching for climate emulation.arXiv preprint arXiv:2512.02268, 2025

    Jeremy Andrew Irvin, Jiaqi Han, Zikui Wang, Abdulaziz Alharbi, Yufei Zhao, Nomin-Erdene Bayarsaikhan, Daniele Visioni, Andrew Y Ng, and Duncan Watson-Parris. Spatiotemporal pyramid flow matching for climate emulation.arXiv preprint arXiv:2512.02268, 2025

    work page arXiv 2025

  55. [55]

    Lord Rayleigh. Lix. on convection currents in a horizontal layer of fluid, when the higher temperature is on the under side.The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, 32(192):529–546, 1916

    work page 1916

  56. [56]

    Steady rayleigh–bénard convection between no-slip boundaries.Journal of Fluid Mechanics, 933:R4, 2022

    Baole Wen, David Goluskin, and Charles R Doering. Steady rayleigh–bénard convection between no-slip boundaries.Journal of Fluid Mechanics, 933:R4, 2022

    work page 2022

  57. [57]

    Diffusion models beat gans on image synthesis

    Prafulla Dhariwal and Alexander Nichol. Diffusion models beat gans on image synthesis. In M. Ranzato, A. Beygelzimer, Y . Dauphin, P.S. Liang, and J. Wortman Vaughan, editors, Advances in Neural Information Processing Systems, volume 34, pages 8780–8794. Curran Associates, Inc., 2021. URL https://proceedings.neurips.cc/paper_files/paper/ 2021/file/49ad23d...

    work page 2021

  58. [58]

    Christopher A. T. Ferro. Fair scores for ensemble forecasts.Quarterly Journal of the Royal Meteorological Society, 140(683):1917–1923, 2014

    work page 1917

  59. [59]

    Orthonormal bases of compactly supported wavelets.Communications on Pure and Applied Mathematics, 41(7):909–996, 1988

    Ingrid Daubechies. Orthonormal bases of compactly supported wavelets.Communications on Pure and Applied Mathematics, 41(7):909–996, 1988. doi: 10.1002/cpa.3160410705

    work page doi:10.1002/cpa.3160410705 1988

  60. [60]

    Lee, Ralf Gommers, Filip Waselewski, Kai Wohlfahrt, and Aaron ;Leary

    Gregory R. Lee, Ralf Gommers, Filip Waselewski, Kai Wohlfahrt, and Aaron ;Leary. Pywavelets: A python package for wavelet analysis.Journal of Open Source Software, 4(36):1237, 2019. doi: 10.21105/joss.01237. URLhttps://doi.org/10.21105/joss.01237

    work page doi:10.21105/joss.01237 2019

  61. [61]

    J., Deems, S., Furlani, T

    Timothy J. Boerner, Stephen Deems, Thomas R. Furlani, Shelley L. Knuth, and John Towns. Access: Advancing innovation: Nsf’s advanced cyberinfrastructure coordination ecosystem: Services & support. InPractice and Experience in Advanced Research Computing 2023: Computing for the Common Good, PEARC ’23, page 173–176, New York, NY , USA, 2023. Association for...

    work page doi:10.1145/3569951.3597559 2023