CHD: Coupled Hierarchical Diffusion for Long-Horizon Tasks

TL;DR

CHD introduces a unified diffusion framework that tightly couples high-level sub-goal selection with low-level trajectory generation, significantly improving long-horizon planning in complex environments.

Contribution

The paper presents Coupled Hierarchical Diffusion (CHD), a novel method that models HL sub-goals and LL trajectories jointly, enabling scalable and coherent long-horizon planning.

Findings

CHD outperforms baseline diffusion methods in maze navigation.

CHD achieves better trajectory coherence in manipulation tasks.

CHD demonstrates scalability across diverse environments.

Abstract

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