This paper tackles a novel problem, extendable long-horizon planning-enabling agents to plan trajectories longer than those in training data without compounding errors. To tackle this, we propose the Hierarchical Multiscale Diffuser (HM-Diffuser) and Progressive Trajectory Extension (PTE), an augmentation method that iteratively generates longer trajectories by stitching shorter ones. HM-Diffuser trains on these extended trajectories using a hierarchical structure, efficiently handling tasks across multiple temporal scales. Additionally, we introduce Adaptive Plan Pondering and the Recursive HM-Diffuser, which consolidate hierarchical layers into a single model to process temporal scales recursively. Experimental results demonstrate the effectiveness of our approach, advancing diffusion-based planners for scalable long-horizon planning.
View on arXiv@article{chen2025_2503.20102,
title={ Extendable Long-Horizon Planning via Hierarchical Multiscale Diffusion },
author={ Chang Chen and Hany Hamed and Doojin Baek and Taegu Kang and Yoshua Bengio and Sungjin Ahn },
journal={arXiv preprint arXiv:2503.20102},
year={ 2025 }
}