HuggingFace Daily Papers（社区热门论文）· HuggingFace Daily Papers（社区热门论文）·· 7 天前AI 评分37
AnisoWM 用各向异性表示改进 JEPA 世界模型规划
AI 导读
针对 JEPA 世界模型中各向同性高斯正则导致的表示几何与任务代价不匹配问题，研究者提出 AnisoWM 与 ΛReg，用固定迹和各向异性约束下的可学习对角协方差替代固定各向同性高斯目标，预测目标、预测器架构与欧氏规划器均保持不变。在四个视觉控制环境中，AnisoWM 的规划成功率均优于 LeWorldModel，其潜在规划代价与任务结果也更一致。
正文
Abstract:Latent world models learn action-conditioned dynamics in representation space and often score candidate actions by Euclidean distance to a goal representation. Joint training typically regularizes the representation to prevent collapse, but the resulting representation geometry also determines how terminal errors are weighted during planning. We show that accurate prediction and noncollapsed representations do not guarantee a task-aligned latent planning cost: isotropic Gaussian regularization can induce a geometry that ranks feasible outcomes differently from the task cost. To address this mismatch, we introduce AnisoWM with $\Lambda$Reg, which replaces the fixed isotropic Gaussian target with a learnable diagonal covariance under fixed-trace and anisotropy constraints. The prediction objective, predictor architecture, and Euclidean planner remain unchanged; the target is used only during training. Our analysis characterizes the prediction-driven allocation of target variance, its dependence on the training distribution, and the conditions under which the induced metric reduces planning regret. Across four visual control environments, AnisoWM improves planning success over LeWorldModel in all four. Its latent planning cost also shows better agreement with task outcomes. Project website: this https URL
| Subjects: | Robotics (cs.RO) |
| Cite as: | arXiv:2609.37441 [cs.RO] |
| (or arXiv:2609.37441v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2609.37441 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Mingu Kang [view email]
[v1]
Tue, 29 Sep 2026 13:00:37 UTC (2,004 KB)
来源：HuggingFace Daily Papers（社区热门论文） · arxiv.org