ComboStoc: Combinatorial Stochasticity for Diffusion Generative Models
ACM Transactions on Graphics · 03.07.2026 · 45 (4) · 1-17
DOI: 10.1145/3811285 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM Transactions on Graphics, online erschienen am 03.07.2026. Verfasst von Rui Xu und 8 weiteren Autor:innen. Die Arbeit nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Start-up fund from Tsinghua University, HKSAR Government under the ITSP-Platform grants.
Abstract (Originalsprache)
In this paper, we study an under-explored but important factor of diffusion generative models, i.e., the combinatorial complexity. Data samples are generally high-dimensional, and for various structured generation tasks, additional attributes are combined to associate with data samples. We show that the space spanned by the combination of dimensions and attributes can be insufficiently covered by existing training schemes of diffusion generative models, potentially limiting test time performance. We present a simple fix to this problem by constructing stochastic processes that fully exploit the combinatorial structures, hence the name ComboStoc. Using this simple strategy, we show that network training is significantly accelerated across diverse data modalities, including images and 3D structured shapes. Moreover, ComboStoc enables a new way of test time generation which uses asynchronous time steps for different dimensions and attributes, thus allowing for varying degrees of control over them. Our code is available at: https://github.com/Xrvitd/ComboStoc.
Förderung
- Start-up fund from Tsinghua University
- HKSAR Government under the ITSP-Platform grants
Zitieren
APA
Xu, R., Wang, J., Pan, H., Liu, Y., Tong, X., Xin, S., Tu, C., Komura, T., & Wang, W. (2026). ComboStoc: Combinatorial Stochasticity for Diffusion Generative Models. ACM Transactions on Graphics, 45(4), 1-17. https://doi.org/10.1145/3811285
Vancouver
Xu R, Wang J, Pan H, Liu Y, Tong X, Xin S, et al. ComboStoc: Combinatorial Stochasticity for Diffusion Generative Models. ACM Transactions on Graphics. 2026;45(4):1-17. doi:10.1145/3811285
BibTeX
@article{Xu2026,
author = {Xu, Rui and Wang, Jiepeng and Pan, Hao and Liu, Yang and Tong, Xin and Xin, Shiqing and Tu, Changhe and Komura, Taku and Wang, Wenping},
title = {ComboStoc: Combinatorial Stochasticity for Diffusion Generative Models},
journal = {ACM Transactions on Graphics},
year = {2026},
volume = {45},
number = {4},
pages = {1--17},
doi = {10.1145/3811285}
}
EUVIMED