让AI像人一样在地图上推理规划，精准调用地理工具。
MapAgent: A Hierarchical Agent for Geospatial Reasoning with Dynamic Map Tool Integration
- 分层设计：先拆解任务再分配专用工具代理，避免信息过载。
- 四类地图评测中表现超越现有模型，尤其在复杂查询上提升显著。
- 适合需要地图交互的智能导航、城市规划等场景开发者使用。
代理型AI通过增强大语言模型（LLM）的复杂推理与工具调用能力，拓展了其应用边界。然而，现有框架多针对数学、编程或网页自动化等任务，难以胜任需空间推理、多跳规划和实时地图交互的地理空间任务。为此，我们提出MapAgent，一种支持动态地图工具集成的分层多代理即插即用框架，配备定制化工具集与代理结构。不同于传统扁平式代理统一处理工具——易因相似但细微差异的地理API导致模型过载——MapAgent将规划与执行分离：高层规划器将复杂查询分解为子目标，并路由至专业化模块。对于地图服务等工具密集型模块，设计专用地图工具代理，可自适应并行调度相关API，高效获取与查询相关的地理空间数据；而简单模块（如答案生成或提取）则无需额外代理开销。该分层架构降低认知负荷，提升工具选择准确率，并实现对相似API的精确协调。我们在四个多样化地理空间基准测试（MapEval-Textual、MapEval-API、MapEval-Visual、MapQA）上评估，结果表明其显著优于当前最先进的工具增强与代理基基线。项目代码已开源：https://github.com/Hasebul/MapAgent。
原文摘要 · Abstract (English)
Agentic AI has significantly extended the capabilities of large language models (LLMs) by enabling complex reasoning and tool use. However, most existing frameworks are tailored to domains such as mathematics, coding, or web automation, and fall short on geospatial tasks that require spatial reasoning, multi-hop planning, and real-time map interaction. To address these challenges, we introduce MapAgent, a hierarchical multi-agent plug-and-play framework with customized toolsets and agentic scaffolds for map-integrated geospatial reasoning. Unlike existing flat agent-based approaches that treat tools uniformly-often overwhelming the LLM when handling similar but subtly different geospatial APIs-MapAgent decouples planning from execution. A high-level planner decomposes complex queries into subgoals, which are routed to specialized modules. For tool-heavy modules-such as map-based services-we then design a dedicated map-tool agent that efficiently orchestrates related APIs adaptively in parallel to effectively fetch geospatial data relevant for the query, while simpler modules (e.g., solution generation or answer extraction) operate without additional agent overhead. This hierarchical design reduces cognitive load, improves tool selection accuracy, and enables precise coordination across similar APIs. We evaluate MapAgent on four diverse geospatial benchmarks-MapEval-Textual, MapEval-API, MapEval-Visual, and MapQA-and demonstrate substantial gains over state-of-the-art tool-augmented and agentic baselines. We open-source our framwork at https://github.com/Hasebul/MapAgent.
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。