Introduction to Flow Matching

Flow matching is a framework for generative modeling that learns a vector field transporting a simple reference distribution to a complex target via an ODE. We introduce the key ideas — conditional probability paths, the continuity equation, and the conditional flow matching loss — and discuss practical extensions including SDE-based samplers, alternative prediction parametrizations, variational flow matching, and classifier-free guidance. We close with a survey of current research directions.

See Reference lecture notes for details.