S-Space：探索多模态语言模型中的空间工作区

S-Space: Exploring Spatial Workspace in Multimodal Language Models

October 9, 2026

作者: Zhiyuan Shi, Xiang Li, Xingzhuo Guo, Shangchen Miao, Weiliang Chen, Qiyu Dai, Jinshan Ren, Jiangran Lyu, Fangfu Liu, Mingsheng Long, Jialong Wu

cs.AI

摘要

多模态语言模型（MLLMs）展现出颇具前景的空间推理能力，但支撑这些能力的内部机制仍鲜为人知。我们揭示了 S-Space：一个线性表示子空间，其中对象 token 表示沿水平、垂直和距离轴编码连续坐标。尽管我们是通过将空间答案得分对对象 token 激活求导来识别这些轴的，但 S-Space 是内在的，而非由特定数据或提示所诱发，并且在非空间问题下仍可读取。我们表明，S-Space 充当 MLLMs 的内部空间工作区。感知将对象位置编码到其中，形成可直接读出的空间表示，以支持准确的空间判断。内部推理会对其进行变换，以适应不断变化的视角。言语响应则有选择地访问它，这一点由因果干预所证明：这些干预会改变空间判断，同时大体上保留非空间响应。案例研究进一步表明，S-Space 会为未见实体和观察者分配位置，并将物理方向与抽象语言含义联系起来，这既体现了大规模语言训练的能力，也体现了其局限。最后，我们将 S-Space 用作诊断空间推理失败的工具。方向词读出揭示了基本方位参考系与图像相对参考系之间的系统性耦合。更重要的是，一种感知-计算分解暴露了一个关键瓶颈：MLLMs 能够形成很强的空间表示，但在跨视角变化进行可靠推理所需的空间变换方面仍有不足。

English

Multimodal language models (MLLMs) exhibit promising spatial reasoning capabilities, yet the internal mechanisms supporting them remain poorly understood. We uncover S-Space, a linear representation subspace in which object-token representations encode continuous coordinates along horizontal, vertical, and distance axes. Although we identify these axes by differentiating spatial-answer scores with respect to object-token activations, S-Space is intrinsic rather than induced by specific data or prompts, remaining readable under non-spatial questions. We show that S-Space functions as an internal spatial workspace of MLLMs. Perception encodes object locations into it, forming spatial representations that can be directly read out to support accurate spatial judgments. Internal reasoning transforms it for changing viewpoints. Verbal responses selectively access it, evidenced by causal interventions that alter spatial judgments while largely preserving non-spatial responses. Case studies further show that S-Space assigns locations to unseen entities and the observer, and connects physical directions with abstract linguistic meanings, suggesting both the capabilities and limitations of large-scale language training. Finally, we use S-Space as a diagnostic tool for spatial reasoning failures. Direction-word readouts reveal systematic coupling between cardinal and image-relative reference frames. More importantly, a perception-computation decomposition exposes a key bottleneck: MLLMs can form capable spatial representations, yet fall short in the spatial transformations required to reason reliably across viewpoint changes.