Extendable Planning via Multiscale Diffusion
This addresses the problem of scalable long-horizon planning for AI systems in complex environments, representing an incremental improvement over existing diffusion-based methods.
The paper tackles the challenge of long-horizon planning in complex environments, where diffusion-based planners are limited by training trajectory lengths, and proposes a two-phase solution with Progressive Trajectory Extension and Hierarchical Multiscale Diffuser, resulting in strong performance gains for scalable and efficient decision-making.
Long-horizon planning is crucial in complex environments, but diffusion-based planners like Diffuser are limited by the trajectory lengths observed during training. This creates a dilemma: long trajectories are needed for effective planning, yet they degrade model performance. In this paper, we introduce this extendable long-horizon planning challenge and propose a two-phase solution. First, Progressive Trajectory Extension incrementally constructs longer trajectories through multi-round compositional stitching. Second, the Hierarchical Multiscale Diffuser enables efficient training and inference over long horizons by reasoning across temporal scales. To avoid the need for multiple separate models, we propose Adaptive Plan Pondering and the Recursive HM-Diffuser, which unify hierarchical planning within a single model. Experiments show our approach yields strong performance gains, advancing scalable and efficient decision-making over long-horizons.