AI看视频的空间感，连人类一半都不到
人类看一段视频，能在大脑里拼出一张俯视图，知道哪里是哪里；最强的AI模型却做不到。研究者造了一个测试：让AI看6个多小时的合成视频，然后回答从没见过的视角下的空间问题，比如某个物体在全局地图的哪个位置。22个顶尖视觉语言模型里，最强的零样本得分42.68，人类是79.08——差了一倍。更关键的是，单独问局部空间问题时AI表现不差，但一旦要把长时间看到的画面整合成一张连贯的全局地图，它就崩了。这不是它看不懂单个画面，而是它没法把碎片拼成整体。它不是你明天能用上的东西，但这条短板决定了AI离真正走进现实世界——比如自己认路、操作物体——还差着关键一步。
📄 原文摘要(英文)
Spatial intelligence is fundamental to embodied agents, yet existing benchmarks focus on local spatial perception from single or few viewpoints, overlooking global spatial awareness over continuous, long-horizon visual streams. To address this limitation, we introduce the Global-Spatial-Temporal Benchmark (GST-Bench), a VQA benchmark for global spatial intelligence in video understanding, comprising human-verified questions derived from 6,790 minutes of synthetically generated video. It requires models to perform accurate spatial inference from novel viewpoints unseen in the input video and to map egocentric observations onto global top-down images. A comprehensive evaluation of 22 state-of-the-art VLMs exposes a striking gap between models and humans: the strongest zero-shot model attains only 42.68, far below the human score of 79.08. To probe the cause of this gap, we construct GST-Bench-Local and find that models, despite strong local spatial understanding under the same task formulation, still fail to consolidate long-horizon observations into a globally consistent scene representation. We further provide GST-Train, a dataset for global spatial reasoning, as a complementary resource to facilitate future research on this challenge.