RAYNOVA：射线空间中的尺度-时序自回归世界建模
RAYNOVA: Scale-Temporal Autoregressive World Modeling in Ray Space
RAYNOVA 提出基于相对 Plücker 射线位置编码的各向同性时空表示，结合双因果自回归框架统一处理尺度与时间拓扑顺序，为多视角视频生成提供了一种无需显式 3D 几何先验的通用方法，可启发视觉研究者设计更鲁棒的世界模型。
原文依据 RAYNOVA constructs an isotropic spatio-temporal representation across views, frames, and scales based on relative Plucker-ray positional encoding
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论世界基础模型旨在以物理上合理的行为模拟真实世界的演化。与先前分别处理空间和时间相关性的方法不同，我们提出 RAYNOVA，一种用于驾驶场景的几何不可知多视角世界模型，采用双因果自回归框架。它在自回归过程中同时遵循尺度维度和时间拓扑顺序，并利用全局注意力进行统一的 4D 时空推理。与施加强 3D 几何先验的现有工作不同，RAYNOVA 基于相对 Plücker 射线位置编码构建跨视角、帧和尺度的各向同性时空表示，从而能够鲁棒地泛化到多样化的相机设置和自运动。我们进一步引入循环训练范式以缓解长时程视频生成中的分布漂移。RAYNOVA 在 nuScenes 上取得了最优的多视角视频生成结果，同时在多样化输入条件下提供更高的吞吐量和强可控性，无需显式 3D 场景表示即可泛化到新视角和相机配置。我们的代码将在 https://raynova-ai.github.io/ 发布。
英文原文摘要 1225 字符
World foundation models aim to simulate the evolution of the real world with physically plausible behavior. Unlike prior methods that handle spatial and temporal correlations separately, we propose RAYNOVA, a geometry-agonistic multiview world model for driving scenarios that employs a dual-causal autoregressive framework. It follows both scale-wise and temporal topological orders in the autoregressive process, and leverages global attention for unified 4D spatio-temporal reasoning. Different from existing works that impose strong 3D geometric priors, RAYNOVA constructs an isotropic spatio-temporal representation across views, frames, and scales based on relative Plucker-ray positional encoding, enabling robust generalization to diverse camera setups and ego motions. We further introduce a recurrent training paradigm to alleviate distribution drift in long-horizon video generation. RAYNOVA achieves state-of-the-art multi-view video generation results on nuScenes, while offering higher throughput and strong controllability under diverse input conditions, generalizing to novel views and camera configurations without explicit 3D scene representation. Our code will be released at https://raynova-ai.github.io/.