Risk-Aware LLM Agents for Geospatial Data Retrieval: Design and Preliminary Adversarial Evaluation
风险感知的LLM智能体用于地理空间数据检索:设计与初步对抗性评估
机构 * Dept. of Systems Design Engineering, University of Waterloo(滑铁卢大学系统设计工程系) ; SkyWatch ; Dept. of Geography and Environmental Management, University of Waterloo(滑铁卢大学地理与环境管理系) ; Dept. of Geomatics Engineering, University of Calgary(卡尔加里大学测绘工程系)
专题命中 机器人数据与评测 :manipulation(abstract);分类 cs.AI
AI总结 提出一种基于LLM的框架,通过自然语言查询从云地理空间目录检索遥感数据,集成三个智能体实现安全、意图解析和API调用生成,初步对抗实验表明提示级安全指令提升鲁棒性但需系统级防御。
Comments Accepted for publication in the International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences (ISPRS Archives), ISPRS Congress 2026
Journal ref Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIX-B3-2026, 1419-1426, 2026