Concept
Conditional flow matching
Conditional flow matching (CFM) is a training technique for generative models that learn to transform noise into data by following a learned velocity field. Think of it like diffusion, except instead of predicting noise across many stochastic denoising steps, you train a…
The rest of “Conditional flow matching” is a premium feature: every concept in the library gets a precise, practitioner-focused write-up like this one, cross-linked straight from the paper summaries that use it.
Log in to unlock→