Inverse Flow and Consistency Models

Yuchen Zhang, Jian Zhou

Abstract: Inverse generation problems, such as denoising without ground truth observations, is a critical challenge in many scientific inquiries and real-world applications. While recent advances in generative models like diffusion models, conditional flow matching, and consistency models achieved impressive results by casting generation as denoising problems, they cannot be directly used for inverse generation without access to clean data. Here we introduce Inverse Flow (IF), a novel framework that enables using these generative models for inverse generation problems including denoising without ground truth. Inverse Flow can be flexibly applied to nearly any continuous noise distribution and allows complex dependencies. We propose two algorithms for learning Inverse Flows, Inverse Flow Matching (IFM) and Inverse Consistency Model (ICM). Notably, to derive the computationally efficient, simulation-free inverse consistency model objective, we generalized consistency training to any forward diffusion processes or conditional flows, which have applications beyond denoising. We demonstrate the effectiveness of IF on synthetic and real datasets, outperforming prior approaches while enabling noise distributions that previous methods cannot support. Finally, we showcase applications of our techniques to fluorescence microscopy and single-cell genomics data, highlighting IF's utility in scientific problems. Overall, this work expands the applications of powerful generative models to inversion generation problems.

Lay Summary: We often get data that’s messy or “noisy”—for example, blurry microscope pictures or shaky readings—without having any perfect examples to learn from. Our new method, called Inverse Flow, treats each noisy example as a step along a path and learns how to go backwards, undoing the noise. Unlike older approaches, it only needs the noisy data itself and doesn’t require any clean reference. We show it can sharpen microscope images and make single-cell gene measurements more accurate, so scientists can uncover hidden details from messy observations in biology, physics, and beyond.

Primary Area: Deep Learning->Generative Models and Autoencoders

Keywords: consistency models, flow matching, diffusion models, inverse problem

Submission Number: 12002