WorldLens：面向真实世界的驾驶世界模型全谱系评测
WorldLens: Full-Spectrum Evaluations of Driving World Models in Real World
为驾驶世界模型提供覆盖生成、重建、动作跟随、下游任务与人类偏好的统一评测协议，并给出可蒸馏的评测模型范式；做视频生成或自动驾驶仿真的研究者可直接用其指标与 WorldLens-26K 标注来诊断模型在几何一致性与物理合理性上的失效模式。
原文依据 a full-spectrum benchmark evaluating how well a model builds, understands, and behaves within its generated world
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论生成式世界模型正在重塑具身智能，使智能体能够合成逼真的四维驾驶环境，这些环境看起来可信，却常常在物理或行为层面失效。尽管进展迅速，该领域仍缺乏统一方式来评估生成的世界是否保持几何、遵循物理规律或支持可靠控制。我们提出 WorldLens，一个全谱系基准，用于评估模型在其生成世界中的构建、理解与行为表现。它涵盖五个方面——生成、重建、动作跟随、下游任务与人类偏好——共同覆盖视觉真实感、几何一致性、物理合理性与功能可靠性。在这些维度上，现有世界模型无一能在所有方面都表现优异：纹理强的模型常违反物理规律，而几何稳定的模型缺乏行为保真度。为使客观指标与人类判断对齐，我们进一步构建 WorldLens-26K，一个带数值评分与文本理由的大规模人工标注视频数据集，并开发 WorldLens-Agent，一个从这些标注中蒸馏得到的评测模型，以实现可扩展、可解释的评分。基准、数据集与评测智能体共同构成衡量世界保真度的统一生态，从而标准化未来模型的评判方式：不仅看它们看起来多真实，也看它们行为得多真实。
英文原文摘要 1380 字符
Generative world models are reshaping embodied AI, enabling agents to synthesize realistic 4D driving environments that look convincing but often fail physically or behaviorally. Despite rapid progress, the field still lacks a unified way to assess whether generated worlds preserve geometry, obey physics, or support reliable control. We introduce **WorldLens**, a full-spectrum benchmark evaluating how well a model builds, understands, and behaves within its generated world. It spans five aspects - Generation, Reconstruction, Action-Following, Downstream Task, and Human Preference - jointly covering visual realism, geometric consistency, physical plausibility, and functional reliability. Across these dimensions, no existing world model excels universally: those with strong textures often violate physics, while geometry-stable ones lack behavioral fidelity. To align objective metrics with human judgment, we further construct **WorldLens-26K**, a large-scale dataset of human-annotated videos with numerical scores and textual rationales, and develop **WorldLens-Agent**, an evaluation model distilled from these annotations to enable scalable, explainable scoring. Together, the benchmark, dataset, and agent form a unified ecosystem for measuring world fidelity - standardizing how future models are judged not only by how real they look, but by how real they behave.