EgoMind：通过语言推理激活多模态大语言模型的空间认知
EgoMind: Activating Spatial Cognition through Linguistic Reasoning in MLLMs
提出一种无需3D几何先验的纯语言思维链框架，通过角色扮演描述构建跨帧语言场景图，并渐进式分析空间问题，为视觉语言模型中的空间推理提供了可复用的提示与训练策略。
原文依据 EgoMind, a Chain-of-Thought framework that enables geometry-free spatial reasoning through Role-Play Caption
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论多模态大语言模型（MLLMs）正日益被应用于空间认知任务，在这些任务中，它们需要理解和与复杂环境交互。现有大多数工作通过引入3D先验或几何监督来提升空间推理能力，这虽然提高了性能，却带来了大量的数据准备与对齐成本。相比之下，纯2D方法由于难以捕捉跨帧的空间关系，往往在多帧空间推理上表现不佳。为解决这些局限，我们提出EgoMind，一个思维链框架，通过角色扮演描述实现无几何的空间推理，该描述跨帧联合构建连贯的语言场景图，并通过渐进式空间分析逐步推理至任务特定问题。仅使用5K自动生成的SFT样本和20K强化学习样本，EgoMind在VSI-Bench、SPAR-Bench、SITE-Bench和SPBench上取得了有竞争力的结果，证明了其在增强MLLMs空间推理能力方面的有效性，并凸显了语言推理用于空间认知的潜力。代码与数据发布于https://github.com/Hyggge/EgoMind。
英文原文摘要 1224 字符
Multimodal large language models (MLLMs) are increasingly being applied to spatial cognition tasks, where they are expected to understand and interact with complex environments. Most existing works improve spatial reasoning by introducing 3D priors or geometric supervision, which enhances performance but incurs substantial data preparation and alignment costs. In contrast, purely 2D approaches often struggle with multi-frame spatial reasoning due to their limited ability to capture cross-frame spatial relationships. To address these limitations, we propose EgoMind, a Chain-of-Thought framework that enables geometry-free spatial reasoning through Role-Play Caption, which jointly constructs a coherent linguistic scene graph across frames, and Progressive Spatial Analysis, which progressively reasons toward task-specific questions. With only 5K auto-generated SFT samples and 20K RL samples, EgoMind achieves competitive results on VSI-Bench, SPAR-Bench, SITE-Bench, and SPBench, demonstrating its effectiveness in strengthening the spatial reasoning capabilities of MLLMs and highlighting the potential of linguistic reasoning for spatial cognition. Code and data are released at https://github.com/Hyggge/EgoMind.