发表机构
NYU; Advanced Machine Intelligence; INRIA Paris; Brown University（纽约大学; 先进机器智能; 巴黎国家信息与自动化研究所; 布朗大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
H-JEPA提出端到端训练层次化JEPA世界模型，通过自上而下的规划与时间分解，在Visual AntMaze上将成功率从18%提升至73%，并扩展到真实机器人视频。
AI 中文摘要
使用潜在世界模型进行长时程规划需要在多个时间尺度和抽象层次上进行推理。现有的任务无关JEPA世界模型在单一时间尺度上预测和规划，或在共享潜在空间中使用多个视界。我们提出H-JEPA，一种端到端的训练层次化动作条件JEPA的方法，其中每个层级在其自身学习的潜在空间中预测更远的未来。规划自上而下进行：顶层优化朝向目标的进展，每个层级的预测成为其下方规划器的子目标。当数据中的因素以分离的时间尺度演化时，较高层级会丢弃快速、不可预测的细节，并保留较慢的与任务相关的状态。在四个模拟导航和操作环境中，层次化规划优于平坦JEPA；在Visual AntMaze上，三层层次结构将成功率从18%提高到73%，同时使用更少的规划器计算。消融实验将这些收益归因于时间分解和更高层级的目标表示。通过逆动力学监督，该方法扩展到来自DROID的多样真实机器人视频，在较低规划器计算下提高了离线规划保真度。
英文摘要
Long-horizon planning with latent world models requires reasoning across timescales and levels of abstraction. Existing task-agnostic JEPA world models predict and plan at a single timescale or with multiple horizons in one shared latent space. We introduce H-JEPA, an end-to-end recipe for training a hierarchy of action-conditioned JEPAs in which each level predicts farther ahead in its own learned latent space. Planning proceeds top-down: the top level optimizes progress toward the goal, and each level's predictions become subgoals for the planner below it. When factors in the data evolve at separated timescales, higher levels discard fast, unpredictable detail and retain slower task-relevant state. Across four simulated navigation and manipulation environments, hierarchical planning improves over a flat JEPA; on Visual AntMaze, a three-level hierarchy raises success from 18% to 73% using less planner compute. Ablations attribute these gains to both temporal decomposition and higher-level goal representations. With inverse-dynamics supervision, the approach extends to diverse real-robot videos from DROID, where hierarchy improves offline planning fidelity at lower planner compute.
