让扩散模型能规划更长轨迹，突破训练长度限制。
Extendable Planning via Multiscale Diffusion
- 分阶段构建长轨迹，通过多轮拼接逐步扩展
- 跨时间尺度推理，实现高效长时序规划
- 单模型统一处理，适合需要长期决策的场景
长时序规划在复杂环境中至关重要，但基于扩散的规划器（如 Diffuser）受限于训练中观察到的轨迹长度。这导致一个矛盾：长轨迹需用于有效规划，却会降低模型性能。本文提出可扩展长时序规划挑战，并设计两阶段解决方案：第一阶段，渐进式轨迹扩展通过多轮组合拼接构建更长轨迹；第二阶段，分层多尺度扩散模型（Hierarchical Multiscale Diffuser）通过跨时间尺度推理实现长时程高效训练与推断。为避免多个独立模型，提出自适应计划权衡与递归型 HM-Diffuser，将分层规划统一于单一模型。实验表明，该方法显著提升性能，推动可扩展、高效的长时序决策发展。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。