D-JEPA：一种决策对齐的潜在世界模型
D-JEPA: A Decision-Aligned Latent World Model
- The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））
- Boston University（波士顿大学）
- Shanghai Jiao Tong University（上海交通大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
D-JEPA提出决策对齐的潜在世界模型，通过学习候选未来间的决策相关关系，弥补预测与执行成功间的差距，在多个控制任务中显著提升动作选择性能。
AI中文摘要：
潜在世界模型预测动作的后果，但准确的预测并不能保证潜在距离反映哪个候选动作将成功执行。我们识别出一个决策局部的预测差距：在少数竞争执行的未来中，一个预测更接近目标的候选动作可能产生比可用替代方案更差的实际结果。我们提出D-JEPA，一种决策对齐的潜在世界模型，它从已执行的结果中学习候选未来之间的决策相关关系。一个有界、置换等变的算子联合推理目标相对预测特征和序数证据，在动作选择最关键的之处细化预训练的预测几何。受限预测器适应和共享序数接口将这种对齐扩展到互补的预测几何。D-JEPA进一步在JEPA兼容的未来表示中实现学习到的决策结构，使得通过原生潜在距离规划进行部署成为可能。在潜在控制、操作、预训练动作生成模型、物理机器人和自动驾驶中的评估表明，动作选择得到改进，包括在PushT上达到87.89%的成功率，在RoboTwin上平均提升15.04个百分点，以及在物理机器人任务上提升17个百分点。这些结果确立了决策相关的关系结构作为预测世界建模与有效控制之间的直接桥梁。
英文摘要：
Latent world models predict the consequences of actions, but accurate prediction does not guarantee that latent distance reflects which candidate will execute successfully. We identify a decision-local prediction gap: among the few futures competing for execution, a candidate predicted closer to the goal can produce a worse realized outcome than an available alternative. We introduce D-JEPA, a decision-aligned latent world model that learns decision-relevant relations among candidate futures from executed outcomes. A bounded, permutation-equivariant operator jointly reasons over goal-relative predictive features and ordinal evidence, refining pretrained predictive geometry where action choices are most consequential. Restricted predictor adaptation and a shared ordinal interface extend this alignment across complementary predictive geometries. D-JEPA further realizes the learned decision structure in JEPA-compatible future representations, enabling deployment through native latent-distance planning. Evaluations across latent control, manipulation, pretrained action-producing models, physical robots and autonomous driving demonstrate improved action selection, including 87.89% success on PushT, a 15.04-point average gain on RoboTwin, and a 17-point gain on physical robot tasks. These results establish decision-relevant relational structure as a direct bridge between predictive world modeling and effective control.
