Lost in Latent Space: An Empirical Study of Latent Diffusion Models for Physics Emulation

• URL: http://arxiv.org/abs/2507.02608v1
• Date: Thu, 03 Jul 2025 13:32:50 GMT
• Title: Lost in Latent Space: An Empirical Study of Latent Diffusion Models for Physics Emulation
• Authors: François Rozet, Ruben Ohana, Michael McCabe, Gilles Louppe, François Lanusse, Shirley Ho,
• Abstract summary: We investigate whether a similar strategy can be effectively applied to the emulation of dynamical systems.<n>We find that the accuracy of latent-space emulation is surprisingly robust to a wide range of compression rates.<n>We also show that diffusion-based emulators are consistently more accurate than non-generative counterparts.
• Score: 10.924642223683634
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The steep computational cost of diffusion models at inference hinders their use as fast physics emulators. In the context of image and video generation, this computational drawback has been addressed by generating in the latent space of an autoencoder instead of the pixel space. In this work, we investigate whether a similar strategy can be effectively applied to the emulation of dynamical systems and at what cost. We find that the accuracy of latent-space emulation is surprisingly robust to a wide range of compression rates (up to 1000x). We also show that diffusion-based emulators are consistently more accurate than non-generative counterparts and compensate for uncertainty in their predictions with greater diversity. Finally, we cover practical design choices, spanning from architectures to optimizers, that we found critical to train latent-space emulators.

Related papers

• Hybrid Neural-MPM for Interactive Fluid Simulations in Real-Time [57.30651532625017]
  We present a novel hybrid method that integrates numerical simulation, neural physics, and generative control.<n>Our system demonstrates robust performance across diverse 2D/3D scenarios, material types, and obstacle interactions.<n>We promise to release both models and data upon acceptance.
  arXiv  Detail & Related papers  (2025-05-25T01:27:18Z)
• GausSim: Foreseeing Reality by Gaussian Simulator for Elastic Objects [55.02281855589641]
  GausSim is a novel neural network-based simulator designed to capture the dynamic behaviors of real-world elastic objects represented through Gaussian kernels.<n>We leverage continuum mechanics and treat each kernel as a Center of Mass System (CMS) that represents continuous piece of matter.<n>In addition, GausSim incorporates explicit physics constraints, such as mass and momentum conservation, ensuring interpretable results and robust, physically plausible simulations.
  arXiv  Detail & Related papers  (2024-12-23T18:58:17Z)
• Text2PDE: Latent Diffusion Models for Accessible Physics Simulation [7.16525545814044]
  We introduce several methods to apply latent diffusion models to physics simulation.<n>We show that the proposed approach is competitive with current neural PDE solvers in both accuracy and efficiency.<n>By introducing a scalable, accurate, and usable physics simulator, we hope to bring neural PDE solvers closer to practical use.
  arXiv  Detail & Related papers  (2024-10-02T01:09:47Z)
• Resistive Memory-based Neural Differential Equation Solver for Score-based Diffusion Model [55.116403765330084]
  Current AIGC methods, such as score-based diffusion, are still deficient in terms of rapidity and efficiency. We propose a time-continuous and analog in-memory neural differential equation solver for score-based diffusion. We experimentally validate our solution with 180 nm resistive memory in-memory computing macros.
  arXiv  Detail & Related papers  (2024-04-08T16:34:35Z)
• Redefining Super-Resolution: Fine-mesh PDE predictions without classical simulations [0.0]
  We propose a novel definition of super-resolution tailored for PDE-based problems. We use coarse-grid simulated data as our input and predict fine-grid simulated outcomes. Our method enables the generation of fine-mesh solutions bypassing traditional simulation.
  arXiv  Detail & Related papers  (2023-11-16T10:13:09Z)
• Complexity Matters: Rethinking the Latent Space for Generative Modeling [65.64763873078114]
  In generative modeling, numerous successful approaches leverage a low-dimensional latent space, e.g., Stable Diffusion. In this study, we aim to shed light on this under-explored topic by rethinking the latent space from the perspective of model complexity.
  arXiv  Detail & Related papers  (2023-07-17T07:12:29Z)
• Temporal Subsampling Diminishes Small Spatial Scales in Recurrent Neural Network Emulators of Geophysical Turbulence [0.0]
  We investigate how an often overlooked processing step affects the quality of an emulator's predictions. We implement ML architectures from a class of methods called reservoir computing: (1) a form of spatial Vector Autoregression (N VAR), and (2) an Echo State Network (ESN) In all cases, subsampling the training data consistently leads to an increased bias at small scales that resembles numerical diffusion.
  arXiv  Detail & Related papers  (2023-04-28T21:34:53Z)
• Bayesian Emulation for Computer Models with Multiple Partial Discontinuities [0.0]
  An emulator is a fast statistical construct that mimics the slow to evaluate computer model. We introduce the TENSE framework, based on carefully designed correlation structures that respect the discontinuities. We apply the TENSE framework to the TNO Challenge II, emulating the OLYMPUS reservoir model.
  arXiv  Detail & Related papers  (2022-10-19T11:14:57Z)
• Deep convolutional surrogates and degrees of freedom in thermal design [0.0]
  Convolutional Neural Networks (CNNs) are used to predict results of Computational Fluid Dynamics (CFD) directly from topologies saved as images. We present surrogate models for heat transfer and pressure drop prediction of complex fin geometries generated using composite Bezier curves.
  arXiv  Detail & Related papers  (2022-08-16T00:45:39Z)
• Physics-informed machine learning with differentiable programming for heterogeneous underground reservoir pressure management [64.17887333976593]
  Avoiding over-pressurization in subsurface reservoirs is critical for applications like CO2 sequestration and wastewater injection. Managing the pressures by controlling injection/extraction are challenging because of complex heterogeneity in the subsurface. We use differentiable programming with a full-physics model and machine learning to determine the fluid extraction rates that prevent over-pressurization.
  arXiv  Detail & Related papers  (2022-06-21T20:38:13Z)
• Designing Air Flow with Surrogate-assisted Phenotypic Niching [117.44028458220427]
  We introduce surrogate-assisted phenotypic niching, a quality diversity algorithm. It allows to discover a large, diverse set of behaviors by using computationally expensive phenotypic features. In this work we discover the types of air flow in a 2D fluid dynamics optimization problem.
  arXiv  Detail & Related papers  (2021-05-10T10:45:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.