Glossary › matching
GLOSSARY
matching
appears in 2 paper titles
Definition
Training a model to agree with a target quantity, typically by regression. In generative modeling it names a family of objectives — score matching, flow matching — where the model is fit to a target score function or velocity field rather than to a likelihood directly, which is how diffusion and flow-based models are usually formulated in current work. The same word also carries its everyday sense of pairing or retrieval, so read the modifier.
Explainers using this term
- Paper explained: Self-OPD — an image generator that distills itself, with no teacherSelf-OPD: On-Policy Distillation for Flow Matching Models without Teacher
- Flow Matching from Scratch — What Came After Diffusion, and Why It Goes StraightFlow Matching for Generative Modeling