aiqbench analysis
This study applies pre-trained Variational Autoencoders from Stable Diffusion to accelerate gyrokinetic plasma turbulence simulations, achieving significant speedups via physics-informed fine-tuning. By leveraging existing image generation models for spatial compression and combining them with ConvLSTM networks, the framework reduces inference time to seconds on a single GPU while maintaining spectral accuracy across scales. The approach demonstrates a transferable methodology for surrogate modeling in complex physical systems without requiring architecture redesigns for new codes.
A Pre-trained Variational Autoencoder for Gyrokinetic Plasma Turbulence Surrogate Modeling
arXiv:2609.38438v1 Announce Type: cross Abstract: Machine learning surrogate models offer a promising path toward accelerating plasma turbulence simulations. We present PreVAE-Turb, a surrogate modeling framework that leverages pre-trained variational autoencoders (VAEs) from the Stable Diffusion image generation model for efficient spatial compression of turbulence fields. The pre-trained VAE is fine-tuned on turbulence data using a physics-informed loss function that includes a spectral loss operating in Fourier space to enforce spectral accuracy across scales. The VAE is combined with convolutional long short-term memory (ConvLSTM) networks to learn temporal dynamics in latent space, with a manifold consistency error metric that monitors encode--decode consistency during autoregressive rollouts. We validate the framework on two-dimensional Hasegawa-Wakatani drift-wave turbulence and extend it to gyrokinetic turbulence from the GENE code, where a four-channel adaptation simultaneously predicts electrostatic potential, density, and parallel/perpendicular temperature fluctuations without requiring architecture redesign. Once trained, inference generates thousands of time steps in seconds on a single GPU, providing substantial computational acceleration compared to direct numerical simulation. The pre-trained approach offers a transferable methodology broadly applicable to various turbulence simulation codes.
Abstract:Machine learning surrogate models offer a promising path toward accelerating plasma turbulence simulations. We present PreVAE-Turb, a surrogate modeling framework that leverages pre-trained variational autoencoders (VAEs) from the Stable Diffusion image generation model for efficient spatial compression of turbulence fields. The pre-trained VAE is fine-tuned on turbulence data using a physics-informed loss function that includes a spectral loss operating in Fourier space to enforce spectral accuracy across scales. The VAE is combined with convolutional long short-term memory (ConvLSTM) networks to learn temporal dynamics in latent space, with a manifold consistency error metric that monitors encode--decode consistency during autoregressive rollouts. We validate the framework on two-dimensional Hasegawa-Wakatani drift-wave turbulence and extend it to gyrokinetic turbulence from the GENE code, where a four-channel adaptation simultaneously predicts electrostatic potential, density, and parallel/perpendicular temperature fluctuations without requiring architecture redesign. Once trained, inference generates thousands of time steps in seconds on a single GPU, providing substantial computational acceleration compared to direct numerical simulation. The pre-trained approach offers a transferable methodology broadly applicable to various turbulence simulation codes.
Plasma Physics (physics.plasm-ph); Machine Learning (cs.LG)
arXiv:2609.38438 [physics.plasm-ph]
(or arXiv:2609.38438v1 [physics.plasm-ph] for this version)
https://doi.org/10.48550/arXiv.2609.38438 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 267 words
Related coverage
GyroSwin: 5D Surrogates for Gyrokinetic Plasma Turbulence Simulations
arXiv · Sep 7, 2026
Patch-Based 3D Variational Autoencoder for Super-Resolution of Turbulent Channel Flow
arXiv · Aug 3, 2026
TokaMind: A Multi-Modal Transformer Foundation Model for Tokamak Plasma Dynamics
arXiv · Jun 8, 2026
Large language models as synthetic clinical experts to inform longitudinal rare-disease modeling
arXiv · Aug 18, 2026