aiqbench analysis
This research adapts the generative drifting framework to fluid mechanics, achieving CFD surrogate accuracy comparable to iterative diffusion methods while operating two orders of magnitude faster. By utilizing a conditional architecture in a VAE latent space, the approach enables single-pass generation critical for real-time indoor environment optimization. The work also introduces spatial conditioning to improve generalization across unseen geometries, offering a significant efficiency gain for high-fidelity flow modeling tasks.
Drifting Models for Surrogate Flow Modeling
arXiv:2606.07481v1 Announce Type: new Abstract: While Computational Fluid Dynamics (CFD) provides high-fidelity flow fields for optimizing indoor environments, its computational cost limits rapid exploration. To solve this problem generative surrogates offer better distribution modeling than deterministic networks, but iterative sampling is slow. To enable high-quality, single-pass generation, we adapt the novel generative drifting framework to fluid mechanics. We introduce a conditional architecture that performs drifting in a learned VAE latent space and uses label-aware masking to align generated samples with their boundary conditions. Our label-conditioned model matches iterative diffusion in accuracy and flow consistency while running two orders of magnitude faster. Additionally, we propose a spatial-conditioning variant that establishes a promising path towards generalization to unseen geometries. Ultimately, conditional drifting serves as a highly efficient alternative to diffusion based approaches, unlocking real-time CFD surrogates where inference speed is critical.
Abstract:While Computational Fluid Dynamics (CFD) provides high-fidelity flow fields for optimizing indoor environments, its computational cost limits rapid exploration. To solve this problem generative surrogates offer better distribution modeling than deterministic networks, but iterative sampling is slow. To enable high-quality, single-pass generation, we adapt the novel generative drifting framework to fluid mechanics. We introduce a conditional architecture that performs drifting in a learned VAE latent space and uses label-aware masking to align generated samples with their boundary conditions. Our label-conditioned model matches iterative diffusion in accuracy and flow consistency while running two orders of magnitude faster. Additionally, we propose a spatial-conditioning variant that establishes a promising path towards generalization to unseen geometries. Ultimately, conditional drifting serves as a highly efficient alternative to diffusion based approaches, unlocking real-time CFD surrogates where inference speed is critical.
Machine Learning (cs.LG)
arXiv:2606.07481 [cs.LG]
(or arXiv:2606.07481v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2606.07481 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
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 · 228 words
Related coverage
A Continual Validation, Updating, and Decision-Making Framework for Self-Adaptive Digital Twins via Robust Model Predictive Control: A Case Study in Additive Manufacturing
arXiv · Jul 21, 2026
LieStoNet: Learning Lie Symmetries from Spatiotemporal Data for Stochastic Dynamical Systems
arXiv · Aug 4, 2026
Mesh-Native Physics-Informed Graph Surrogates for TCAD-in-the-Loop Design Space Exploration
arXiv · Sep 4, 2026
Diffusion Policy Optimization without Drifting Apart
arXiv · Jun 15, 2026