给AI装上「空间想象力」：看不见的也能猜
现在的AI看图说话很厉害，但一旦要它想象「从背面看这个物体是什么样」或者「被挡住的路怎么走」，它就抓瞎了——因为信息不在画面里。研究者发明了一种叫「想象感知令牌」的东西：让AI在内部先画一张「如果我从另一个角度看会看到什么」的草图，再基于这张草图推理。在三个新测试（视角转换、路径追踪、多视角计数）上，这种方法比让AI用文字一步步推理更准，甚至有时文字推理反而会拖后腿。它不是你明天能用上的功能，但揭示了AI空间推理的一个新方向：与其用语言绕弯子，不如让AI直接「脑补」画面。
📄 原文摘要(英文)
Vision language models (VLMs) excel at many tasks but still struggle with spatial reasoning when critical information is not directly observable. Many such problems require imaginative perception: inferring what would be seen from an unseen viewpoint, tracing paths through occluded spaces, or integrating partial observations into a coherent spatial representation. We introduce Imaginative Perception Tokens (IPT), intermediate perceptual representations that externalize what a VLM would perceive under alternative spatial configurations while remaining consistent with the observed input. To study this capability, we formulate three tasks, Perspective Taking (PET), Path Tracing (PT), and Multiview Counting (MVC), and construct datasets of approximately 20K examples with ground truth imaginations, answers, and evaluation benchmarks. Using the unified VLM BAGEL as the backbone, IPT supervision consistently improves spatial reasoning and often outperforms textual chain of thought training, even without generating images at inference time. On MVC, IPT improves accuracy by 3.4% and achieves competitive performance with strong closed-source models on PT. We further find that combining IPT and label-only supervision yields additional gains, whereas textual chain of thought can substantially degrade performance, suggesting a modality mismatch when spatial computation is forced through language. Overall, IPT provides a principled supervision signal for reasoning about unobserved spatial structure, improving generalization while producing interpretable intermediate representations.