U4D：面向激光雷达序列的不确定性感知 4D 世界建模
U4D: Uncertainty-Aware 4D World Modeling from LiDAR Sequences
用预训练分割模型输出的空间不确定性图来引导扩散生成，先重建高熵难区再补全其余区域，并配 MoST 块融合时空表征——这套「不确定性引导的由难到易生成」可直接迁移到点云/占据栅格/视频等任何需要时序一致性的生成式世界模型。
原文依据 estimates spatial uncertainty maps from a pretrained segmentation model to localize semantically challenging regions
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论从激光雷达序列建模动态三维环境，是构建面向自动驾驶与具身智能的可靠 4D 世界的核心。然而，现有生成式框架往往对所有空间区域一视同仁，忽略了真实场景中各处不确定性的差异。这种均匀生成会在复杂或歧义区域产生伪影，限制了真实感与时序稳定性。本文提出 U4D，一个不确定性感知的 4D 激光雷达世界建模框架。该方法首先从预训练分割模型估计空间不确定性图，以定位语义上具有挑战性的区域；随后通过两个顺序阶段以「由难到易」的方式执行生成：(1) 不确定性区域建模，以精细的几何保真度重建高熵区域；(2) 不确定性条件补全，在学习到的结构先验下合成其余区域。为进一步保证时序一致性，U4D 引入时空混合（MoST）模块，在扩散过程中自适应融合空间与时序表征。大量实验表明，U4D 能生成几何忠实且时序一致的激光雷达序列，提升了面向自动驾驶感知与仿真的 4D 世界建模的可靠性。
英文原文摘要 1311 字符
Modeling dynamic 3D environments from LiDAR sequences is central to building reliable 4D worlds for autonomous driving and embodied AI. Existing generative frameworks, however, often treat all spatial regions uniformly, overlooking the varying uncertainty across real-world scenes. This uniform generation leads to artifacts in complex or ambiguous regions, limiting realism and temporal stability. In this work, we present **U4D**, an uncertainty-aware framework for 4D LiDAR world modeling. Our approach first estimates spatial uncertainty maps from a pretrained segmentation model to localize semantically challenging regions. It then performs generation in a "hard-to-easy" manner through two sequential stages: (1) *uncertainty-region modeling*, which reconstructs high-entropy regions with fine geometric fidelity, and (2) *uncertainty-conditioned completion*, which synthesizes the remaining areas under learned structural priors. To further ensure temporal coherence, U4D incorporates a mixture of spatio-temporal (MoST) block that adaptively fuses spatial and temporal representations during diffusion. Extensive experiments show that U4D produces geometrically faithful and temporally consistent LiDAR sequences, advancing the reliability of 4D world modeling for autonomous perception and simulation.