RESEARCH PAPERYear 2024

One-step Diffusion with Distribution Matching Distillation

Tianwei Yin; Michaël Gharbi; Richard Zhang; Eli Shechtman; Frédo Durand; William T. Freeman; Taesung Park

Classification

Major category: Foundational work
Quadrant: Not applicable
Architecture: Not applicable
Prediction paradigm: Not applicable
Subcategories: Training optimization & distillation Diffusion & flow-matching foundations
Source review status: Verified from primary sources

Category review. DMD develops distribution-matching gradients and teacher-pair regression to distill image diffusion into one-step generators. This is pre-2026 generative/distillation methodology, not a WAM model component merely because later WAMs adopt it. Reading evidence

AT A GLANCE

Contribution

A contribution summary has not been added yet.

Abstract

An abstract has not been added yet.

Affiliations

Not listed in the collection.

BibTeX

@inproceedings{dmd,
  title={One-step diffusion with distribution matching distillation},
  author={Yin, Tianwei and Gharbi, Micha{\"e}l and Zhang, Richard and Shechtman, Eli and Durand, Fredo and Freeman, William T and Park, Taesung},
  booktitle={CVPR},
  pages={6613--6623},
  year={2024}
}

Classification

Contribution

Abstract

Affiliations

BibTeX

More in this category

Critique of World Model

Precise and Dexterous Robotic Manipulation via Human-in-the-Loop Reinforcement Learning

Mean Flows for One-step Generative Modeling