On Distilling Generator Matching Models
Abstract
Generator Matching (GM) is a new framework which encompasses the current workhorse generative modeling methods. However GM suffers from the computationally intensive sampling process common to these ODE/SDE based models. We introduce "Implicit Generator Matching" (IGM), a general framework for one-step distillation of generator matching models. Our method generalizes the recently proposed one-step diffusion distillation \citep{zhou2024score,luo2024one} methods to Generator Matching. We present promising initial results on image generation.
Cite
Text
Shankar. "On Distilling Generator Matching Models." ICLR 2025 Workshops: DeLTa, 2025.Markdown
[Shankar. "On Distilling Generator Matching Models." ICLR 2025 Workshops: DeLTa, 2025.](https://mlanthology.org/iclrw/2025/shankar2025iclrw-distilling/)BibTeX
@inproceedings{shankar2025iclrw-distilling,
title = {{On Distilling Generator Matching Models}},
author = {Shankar, Shiv},
booktitle = {ICLR 2025 Workshops: DeLTa},
year = {2025},
url = {https://mlanthology.org/iclrw/2025/shankar2025iclrw-distilling/}
}