# Drifting Models for Surrogate Flow Modeling

arXiv: https://arxiv.org/abs/2606.07481 · HTML: https://mlsanity.com/paper/2606.07481/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 11 references that are in the map.
- First posted on arXiv 5 Jun 2026.

## Builds on (most cited first, 9 of 9)

- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Fourier Neural Operator for Parametric Partial Differential Equations](https://mlsanity.com/paper/2010.08895/) (2020)
- [Physics vs Distributions: Pareto Optimal Flow Matching with Physics Constraints](https://mlsanity.com/paper/2506.08604/) (2025)
- [Sinkhorn-Drifting Generative Models](https://mlsanity.com/paper/2603.12366/) (2026)
- [Deep Equilibrium Based Neural Operators for Steady-State PDEs](https://mlsanity.com/paper/2312.00234/) (2023)
- [FluidFlow: a flow-matching generative model for fluid dynamics surrogates on unstructured meshes](https://mlsanity.com/paper/2604.08586/) (2026)
- [Operator learning for energy-efficient building ventilation control with computational fluid dynamics simulation of a real-world classroom](https://mlsanity.com/paper/2504.21243/) (2025)
- [Reducing Experimental Testing in Space Propulsion Film Cooling Analyses by Pixelwise Generative Image Interpolation](https://mlsanity.com/paper/2605.29911/) (2026)

## Abstract (arXiv)

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.