Method · Diffusion-based synthesis
Distribution shift in diffusion DD
Mitigating the Distribution Shift of Diffusion-based Dataset Distillation
Yue Xu, Chenyu Hu, Pengyu An, Yong-Lu Li
CVPR 2026 · first public 2026-01
In one paragraph
Identifies two distribution shifts in diffusion-based distillation: an optimal synthetic distribution should simplify, not replicate, the real data distribution given the synthetic set's small capacity, and the sampling process itself introduces a further deviation from the learned distribution; fixes both with an L1 sparsity regularizer during diffusion training and, at sampling time, a synchronous (non-sequential) denoising of the whole synthetic dataset with distribution regularizers.
Where it sits
- Diffusion-based synthesis (Generative priors)
- Setting: Image classification
Design choices
| Prior / networks used | diffusion |
| Optimization regime | generator-fine-tuning |
BibTeX (generated; prefer the venue's official entry)
@article{xu2026mitigating,
title = {Mitigating the Distribution Shift of Diffusion-based Dataset Distillation},
author = {Yue Xu and Chenyu Hu and Pengyu An and Yong-Lu Li},
journal = {CVPR 2026},
year = {2026}
}Nearby in Diffusion-based synthesis
Learnability-guided diffusion — Learnability-Guided Diffusion for Dataset Distillation
Jeffrey A. Chan-Santiago, Mubarak Shah · CVPR 2026notablepaper ↗
ManifoldGD — ManifoldGD: Training-Free Hierarchical Manifold Guidance for Diffusion-Based Dataset Distillation
Ayush Roy, Wei-Yang Alex Lee, Rudrasis Chakraborty et al. · CVPR 2026notablepaper ↗code ↗