Flow Matching Guide and Code

By Yaron Lipman, Marton Havasi, Peter Holderrieth, Neta Shaul, Matt Le, Brian Karrer, Ricky T. Q. Chen, David Lopez-Paz, Heli Ben-Hamu, Itai Gat. arXiv.org, 2024

298 citations recorded by the source corpus.

Flow Matching (FM) is a recent framework for generative modeling that has achieved state-of-the-art performance across various domains, including image, video, audio, speech, and biological structures. This guide offers a comprehensive and self-contained review of FM, covering its mathematical foundations, design choices, and extensions. By also providing a PyTorch package featuring relevant examples (e.g., image and text generation), this work aims to serve as a resource for both novice and experienced researchers interested in understanding, applying and further developing FM.

Subfields: Computer Networks and Communications, Management Science and Operations Research

View Paper: https://www.semanticscholar.org/paper/e3c8cb1e6a111442965bb7d469f5c0e6789f5c40

Citations in the corpus

• OOD Detection for EEG-based Machine Learning in High-Risk Environments (2026, unknown): “We use Euler’s method (with 100 timesteps), implemented in the library by Lipman et al. [24], as ODE solver to obtain the likelihoods and noise samples required by the different methods.”