Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Physical Fidelity Reconstruction via Improved Consistency-Distilled Flow Matching for Dynamical Systems

About

Reconstructing high-fidelity flow fields from low-fidelity observations is a central problem in scientific machine learning, yet recent diffusion and flow-matching models typically rely on iterative sampling, making them costly for latency-sensitive workflows such as ensemble forecasting, real-time visualization, and simulation-in-the-loop inference. We study whether a high-fidelity flow-matching generative model can be compressed into a compact one-step model for fast scientific flow reconstruction. Our approach distills an optimal-transport flow-matching teacher into a one-step consistency model. Low-fidelity observations are incorporated at inference by initializing the generative trajectory from a noised observation along the transport path, allowing an unconditional high-fidelity flow model to perform conditional reconstruction without retraining the teacher. We evaluate this distillation strategy on three fluid benchmarks, Smoke Buoyancy, Turbulent Channel Flow, and Kolmogorov Flow, using coarse-to-fine reconstruction as a controlled testbed at field sizes up to $256 \times 256$. Across these settings, the distilled student retains similar performance of the teacher's model on spectrum metrics, while using roughly half as many parameters and achieving a $12\times$ inference speedup over the flow-matching teacher. Under the same training budget, the distilled student also outperforms a one-step consistency model trained directly from scratch by $23.1\%$ in SSIM, showing that teacher distillation improves training efficiency rather than merely accelerating sampling. These results suggest a promising route for turning future high-capacity scientific generative models into compact reconstruction models that are faster to train, cheaper to run, and easier to deploy.

Sicheng Ma, Tianyue Yang, Xiuzhe Wu, Xiao Xue• 2026

Related benchmarks

TaskDatasetResultRank
Autoregressive Super-Resolution (64 to 256)KF256 (2D Kolmogorov Flow) (test)
RL20.287
13
Super-ResolutionSmoke Buoyancy 32 → 128
RL2 Score0.19
7
Super-ResolutionTurbulent Channel Flow 64 -> 192 super-resolution
RL20.0173
7
Showing 3 of 3 rows

Other info

Follow for update