LeWAM：基于JEPAs的端到端世界与动作建模
LeWAM: End-to-End World and Action Modeling with JEPAs
AI总结：
LeWAM是一种端到端JEPA控制模型，通过联合训练目标与潜在状态约束，在接触丰富、长 horizon 等任务上大幅提升规划成功率与速度，在行为克隆任务中也具竞争力。
摘要：
联合嵌入预测架构（JEPAs）无需像素重建即可学习紧凑的潜在世界模型。现有JEPA流程通常先训练视觉表征和动作条件动力学，再添加控制策略或求解器。我们提出LeWAM，一种用于控制的端到端JEPA，具备：（1）针对下一潜在状态预测与动作流匹配的联合训练目标；（2）对潜在状态的各向同性高斯约束，使单个模型可通过评估采样动作的学习潜在动力学，支持直接策略执行与高效规划。与现有视觉规划领域的JEPAs相比，LeWAM在接触丰富任务上将成功率从28.6%提升至89.7%，在长 horizon 任务上从10.9%提升至46.2%，规划速度提升32.3倍；在仿真与物理机器人的行为克隆任务中，LeWAM也与成熟方法具有竞争力。可解释性分析显示，LeWAM能学习与控制相关的表征和动作分布，有效指导测试时的规划与控制。
英文摘要：
Joint-embedding predictive architectures (JEPAs) learn compact latent world models without pixel reconstruction. Existing JEPA pipelines typically train visual representations and action-conditioned dynamics before adding a control policy or solver. We introduce LeWAM, an end-to-end JEPA for control with: (1) a joint training objective over next latent state prediction and action flow matching, and (2) an isotropic Gaussian constraint on the latent state, enabling a single model to support direct policy execution and efficient planning by evaluating sampled actions with its learned latent dynamics. Compared with prior JEPAs in visual planning, LeWAM improves success rate from 28.6% to 89.7% on contact-rich tasks and from 10.9% to 46.2% on long-horizon tasks, and plans 32.3 times faster. On behavior cloning tasks in simulation and physical robots, LeWAM is also competitive with established methods. Interpretability analyses show that LeWAM learns control-relevant representations and action distributions effectively guiding test-time planning and control.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
