空间记忆代理：基于经验的空间智能程序记忆框架
原标题：Spatial Memory Agent: Experience-Grounded Procedure Memory for Spatial Intelligence
AI 摘要
Spatial Memory Agent (SMA) 提出了一种无需参数更新的框架，通过经验驱动的自我进化提升冻结视觉语言模型的空间推理能力。SMA 在可验证环境中利用验证器引导的反思，将空间经验提炼为可迁移的教训，并分配转移可靠性评分以优化检索。在五个空间基准和四个基础模型上，SMA 取得了最佳或接近最佳的性能，展示了无需外部工具的参数更新自由路径。
正文节选
Spatial Memory Agent: Experience-Grounded Procedure Memory for Spatial Intelligence Abstract A frozen vision-language model improves spatial reasoning by self-evolving through verified experience, reflection, and reusable memory retrieval without parameter updates or external tools. Spatial intelligence is becoming a foundation for embodied agents, robotic planning, and multimodal assistants. To improve the spatial reasoning ability of VLM agents, existing work has mainly followed two lines. One