Saved in:
Bibliographic Details
Main Authors: Hao, Ce, Xiao, Anxing, Xue, Zhiwei, Soh, Harold
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.07261
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911205145182208
author Hao, Ce
Xiao, Anxing
Xue, Zhiwei
Soh, Harold
author_facet Hao, Ce
Xiao, Anxing
Xue, Zhiwei
Soh, Harold
contents Diffusion-based planners have shown strong performance in short-horizon tasks but often fail in complex, long-horizon settings. We trace the failure to loose coupling between high-level (HL) sub-goal selection and low-level (LL) trajectory generation, which leads to incoherent plans and degraded performance. We propose Coupled Hierarchical Diffusion (CHD), a framework that models HL sub-goals and LL trajectories jointly within a unified diffusion process. A shared classifier passes LL feedback upstream so that sub-goals self-correct while sampling proceeds. This tight HL-LL coupling improves trajectory coherence and enables scalable long-horizon diffusion planning. Experiments across maze navigation, tabletop manipulation, and household environments show that CHD consistently outperforms both flat and hierarchical diffusion baselines. Our website is: https://sites.google.com/view/chd2025/home
format Preprint
id arxiv_https___arxiv_org_abs_2505_07261
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CHD: Coupled Hierarchical Diffusion for Long-Horizon Tasks
Hao, Ce
Xiao, Anxing
Xue, Zhiwei
Soh, Harold
Robotics
Artificial Intelligence
Diffusion-based planners have shown strong performance in short-horizon tasks but often fail in complex, long-horizon settings. We trace the failure to loose coupling between high-level (HL) sub-goal selection and low-level (LL) trajectory generation, which leads to incoherent plans and degraded performance. We propose Coupled Hierarchical Diffusion (CHD), a framework that models HL sub-goals and LL trajectories jointly within a unified diffusion process. A shared classifier passes LL feedback upstream so that sub-goals self-correct while sampling proceeds. This tight HL-LL coupling improves trajectory coherence and enables scalable long-horizon diffusion planning. Experiments across maze navigation, tabletop manipulation, and household environments show that CHD consistently outperforms both flat and hierarchical diffusion baselines. Our website is: https://sites.google.com/view/chd2025/home
title CHD: Coupled Hierarchical Diffusion for Long-Horizon Tasks
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2505.07261