Saved in:
Bibliographic Details
Main Authors: Lee, Hyunjun, Lee, Hyunsoo, Han, Sookwan
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.21555
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916773683527680
author Lee, Hyunjun
Lee, Hyunsoo
Han, Sookwan
author_facet Lee, Hyunjun
Lee, Hyunsoo
Han, Sookwan
contents There have been many attempts to leverage multiple diffusion models for collaborative generation, extending beyond the original domain. A prominent approach involves synchronizing multiple diffusion trajectories by mixing the estimated scores to artificially correlate the generation processes. However, existing methods rely on naive heuristics, such as averaging, without considering task specificity. These approaches do not clarify why such methods work and often produce suboptimal results when a heuristic suitable for one task is blindly applied to others. In this paper, we present a probabilistic framework for analyzing why diffusion synchronization works and reveal where heuristics should be focused; modeling correlations between multiple trajectories and adapting them to each specific task. We further identify optimal correlation models per task, achieving better results than previous approaches that apply a single heuristic across all tasks without justification.
format Preprint
id arxiv_https___arxiv_org_abs_2503_21555
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SyncSDE: A Probabilistic Framework for Diffusion Synchronization
Lee, Hyunjun
Lee, Hyunsoo
Han, Sookwan
Machine Learning
Computer Vision and Pattern Recognition
Graphics
There have been many attempts to leverage multiple diffusion models for collaborative generation, extending beyond the original domain. A prominent approach involves synchronizing multiple diffusion trajectories by mixing the estimated scores to artificially correlate the generation processes. However, existing methods rely on naive heuristics, such as averaging, without considering task specificity. These approaches do not clarify why such methods work and often produce suboptimal results when a heuristic suitable for one task is blindly applied to others. In this paper, we present a probabilistic framework for analyzing why diffusion synchronization works and reveal where heuristics should be focused; modeling correlations between multiple trajectories and adapting them to each specific task. We further identify optimal correlation models per task, achieving better results than previous approaches that apply a single heuristic across all tasks without justification.
title SyncSDE: A Probabilistic Framework for Diffusion Synchronization
topic Machine Learning
Computer Vision and Pattern Recognition
Graphics
url https://arxiv.org/abs/2503.21555