English

Projected Coupled Diffusion for Test-Time Constrained Joint Generation

Machine Learning 2026-04-21 v3

Abstract

Modifications to test-time sampling have emerged as an important extension to diffusion algorithms, with the goal of biasing the generative process to achieve a given objective without having to retrain the entire diffusion model. However, generating jointly correlated samples from multiple pre-trained diffusion models while simultaneously enforcing task-specific constraints without costly retraining has remained challenging. To this end, we propose Projected Coupled Diffusion (PCD), a novel test-time framework for constrained joint generation. PCD introduces a coupled guidance term into the generative dynamics to encourage coordination between diffusion models and incorporates a projection step at each diffusion step to enforce hard constraints. Empirically, we demonstrate the effectiveness of PCD in application scenarios of image-pair generation, object manipulation, and multi-robot motion planning. Our results show improved coupling effects and guaranteed constraint satisfaction without incurring excessive computational costs.

Keywords

Cite

@article{arxiv.2508.10531,
  title  = {Projected Coupled Diffusion for Test-Time Constrained Joint Generation},
  author = {Hao Luan and Yi Xian Goh and See-Kiong Ng and Chun Kai Ling},
  journal= {arXiv preprint arXiv:2508.10531},
  year   = {2026}
}

Comments

ICLR 2026. OpenReview: https://openreview.net/forum?id=1FEm5JLpvg. Code: https://github.com/EdmundLuan/pcd

R2 v1 2026-07-01T04:49:40.872Z