Inverse Flow and Consistency Models

TL;DR

This paper introduces Inverse Flow, a flexible framework that adapts generative models for inverse problems like denoising without ground truth, demonstrating superior performance on scientific datasets.

Contribution

The paper presents Inverse Flow and two algorithms, IFM and ICM, enabling generative models to solve inverse problems without ground truth data, expanding their application scope.

Findings

Outperforms prior methods on synthetic and real datasets.

Supports complex noise distributions beyond previous methods.

Demonstrates utility in scientific applications like microscopy and genomics.

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.