Skip to content

Consistency-Distilled Flow Matching

Home / Concepts / Consistency-Distilled Flow Matching

Consistency-Distilled Flow Matching

Auto-generated stub. Edit this file to add more details.

A distillation framework that compresses optimal-transport flow-matching teacher models into one-step consistency models for efficient scientific flow reconstruction.

Why It Matters

This represents the primary methodology for enabling high-fidelity, one-step reconstruction for dynamical systems, which is the core contribution of the paper.

Evidence

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.

Metadata & Links