大模型在迷宫任务中表现依赖输入格式，无法构建稳定空间认知。
Do LLMs Build Spatial World Models? Evidence from Grid-World Maze Tasks
- 用链式思考提示测试大模型迷宫推理能力，对比不同表示形式
- 文本网格下准确率80-86%，视觉网格下骤降至16-34%，差距达2-5倍
- 虽能理解空间语义，却无法累积知识做一致空间计算，适合特定场景
基础模型在多样任务中表现出色，但其构建内部空间世界模型以进行推理与规划的能力仍不明确。我们通过迷宫任务系统评估大语言模型的空间理解能力，该任务需多步规划与空间抽象。在Gemini-2.5-Flash、GPT-5-mini、Claude-Haiku-4.5和DeepSeek-Chat上进行的综合实验显示，模型在空间推理方面存在显著差异。使用思维链提示时，Gemini在5x5至7x7小迷宫上以分词邻接表示获得80-86%准确率，但在视觉网格格式下性能骤降至16-34%，差距达2-5倍，表明其推理具有表示依赖性而非格式不变性。进一步通过顺序邻近性问题与组合距离比较探测空间理解。尽管推理轨迹实现96-99%语义覆盖率，模型未能利用该理解进行一致的空间计算，说明其将每个问题独立处理而非积累空间知识。基于迷宫任务的发现表明，大模型未发展出稳健的空间世界模型，而是表现出依赖表示与提示的推理，仅在狭窄条件下有效。这些结果对需要空间抽象的应用部署具有关键影响。
原文摘要 · Abstract (English)
Foundation models have shown remarkable performance across diverse tasks, yet their ability to construct internal spatial world models for reasoning and planning remains unclear. We systematically evaluate the spatial understanding of large language models through maze tasks, a controlled testing context requiring multi-step planning and spatial abstraction. Across comprehensive experiments with Gemini-2.5-Flash, GPT-5-mini, Claude-Haiku-4.5, and DeepSeek-Chat, we uncover significant discrepancies in spatial reasoning that challenge assumptions about LLM planning capabilities. Using chain-of-thought prompting, Gemini achieves 80-86% accuracy on smaller mazes (5x5 to 7x7 grids) with tokenized adjacency representations, but performance collapses to 16-34% with visual grid formats, which is a 2-5x difference, suggesting representation-dependent rather than format-invariant spatial reasoning. We further probe spatial understanding through sequential proximity questions and compositional distance comparisons. Despite achieving 96-99% semantic coverage in reasoning traces, models fail to leverage this understanding for consistent spatial computations, indicating that they treat each question independently rather than building cumulative spatial knowledge. Our findings based on the maze-solving tasks suggest that LLMs do not develop robust spatial world models, but rather exhibit representation-specific and prompting-dependent reasoning that succeeds only under narrow conditions. These results have critical implications for deploying foundation models in applications requiring spatial abstraction.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。