$Σ$-LeWM：用 S-信息学习世界模型以辅助机器人控制
$Σ$-LeWM: Learning World Models via S-Information for Robot Control
做机器人控制或世界模型的研究者可以看看这篇。它在 LeWorldModel 基础上加了反向动力学和逆动力学两个损失项，规划效率有提升，思路和 JEPA 那条无解码器路线不一样，实验是在仿真里做的。
一篇 arXiv 论文提出 $Σ$-LeWM，针对 LeWorldModel 这类无解码器世界模型，在其潜在前向动力学预测器和熵稳定项之外，新增反向动力学预测器和逆动力学动作预测器两个潜在损失项。
这两个额外损失为潜在嵌入施加了更强的结构约束，论文称由此带来性能和规划效率的进一步提升，并通过 S-信息（S-Information）这一理论框架解释其机制。规划效率的提升被作者视为将模型基控制用于真实机器人实时控制的关键性质。
实验在标准世界模型基准和两个仿真机器人操作任务上进行验证。相比 JEPA 与 LeWorldModel 等仅依赖熵正则的无解码器方案，该方法通过多组预测器组合而非解码器来稳定表示学习。
$Σ$-LeWM: Learning World Models via S-Information for Robot Control
World models have increasingly been applied as a model-based control method for robotics tasks such as manipulation and locomotion. While early world model architectures relied on decoders to prevent mode collapse and learn a predictable latent space, recent works on world models, such as Joint Embedding Predictive Architecture or LeWorldModel, increasingly aim to learn embeddings without a grounding decoder, utilising only a latent forward dynamics predictor and a stabilising entropy term. In this paper, we set out to demonstrate that the further introduction of two additional latent losses to LeWorldModel, a backward dynamics predictor and an inverse dynamics action predictor, enforces further structure on the latent embedding, leading to further gains in both performance and planning efficiency. These are critical properties for improving the applicability of such methods to real-time model-based control of real-world robots. We demonstrate these gains on standard world model benchmarks and two simulated robot manipulation tasks.
本条情报归属 AITOP「论文」核心追踪栏目。在跨区域产业观察坐标中，该动态直接关联「AI 开发者工具链、Coding Agent 与工程化实践」研究主线，反映出当前技术演进与商业落地的最新风向。涉及核心实体与技术议题包括：世界模型、LeWorldModel、JEPA、机器人控制，读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。
信源核验说明：本文原始线索采编自「arXiv cs.AI」（@Xinrui Zu, Michael Przystupa, Zhao Yang, Shujian Yu, Kevin Sebastian Luck），经 AITOP 多维评分管线结构化提炼。
