地理空间信息2025，Vol.23Issue(12)：1-6,6.DOI:10.3969/j.issn.1672-4623.2025.12.001
基于工具集增强的地图制图大模型构建方法研究
Research on the Construction Method of Cartography Large Models Based on Tool-enhanced
摘要
Abstract
Cartography has long faced challenges such as high professional thresholds and low automation levels.We used the tool-enhanced large language model,model fine-tuning,and Web cartography technologies to explore a cartography large models construction method.Through the automated construction of tool sets based on large models,generation of fine-tuning samples,and domain-specific fine-tuning,we developed a pro-totype large language model system for cartography.Experimental results demonstrate that this method can effectively understand users'natural language instructions,accurately call upon relevant cartographic tools,and achieve intelligent map generation.This research provides a new tech-nical approach for intelligent cartography and offers fresh insights into the application of large language models in specialized professional fields.关键词
地图制图/大语言模型/工具增强/模型微调Key words
cartography/large language model/tool-enhanced/model fine-tuning分类
天文与地球科学引用本文复制引用
任东宇,韦钰华,刘阳杰,王叶兰,狄威..基于工具集增强的地图制图大模型构建方法研究[J].地理空间信息,2025,23(12):1-6,6.基金项目
四川省地理时空大数据中心建设关键技术研究项目. （）
