https://ankky-zhao.github.io/zh/tags/llm/LLMHugoBlox Kit (https://hugoblox.com)zh-HansMon, 03 Nov 2025 00:00:00 +0000https://ankky-zhao.github.io/media/icon_hu_1c0e9cb08cfb822a.pnghttps://ankky-zhao.github.io/zh/tags/llm/https://ankky-zhao.github.io/zh/publications/geo-class/Mon, 03 Nov 2025 00:00:00 +0000https://ankky-zhao.github.io/zh/publications/geo-class/<p>网络上共享的大量 Google Earth Engine 脚本包含了地理空间分析的实践知识，涵盖数据选择、云掩膜、指数计算到结果导出等环节。这些知识分散在代码中，难以复用。</p> <p> <figure > <div class="flex justify-center "> <div class="w-full" > <img alt="从众包脚本到地学处理知识图谱" srcset="https://ankky-zhao.github.io/zh/publications/geo-class/geo-class-concept_hu_a41248da9cfbb0d.webp 320w, https://ankky-zhao.github.io/zh/publications/geo-class/geo-class-concept_hu_521aba9b388ba36d.webp 480w, https://ankky-zhao.github.io/zh/publications/geo-class/geo-class-concept_hu_93f1b53097c20be1.webp 760w" sizes="(max-width: 480px) 100vw, (max-width: 768px) 90vw, (max-width: 1024px) 80vw, 760px" src="https://ankky-zhao.github.io/zh/publications/geo-class/geo-class-concept_hu_a41248da9cfbb0d.webp" width="760" height="241" loading="lazy" data-zoomable /></div> </div></figure> <em>从脚本中抽取主题、功能、数据源以及时空范围，构建知识图谱。</em></p> <p>Geo-CLASS 协同大语言模型与小语言模型。在领域模式与模式感知的思维链提示引导下，由大语言模型识别脚本中的实体与关系；再由小语言模型通过语义向量化、余弦相似度与最佳匹配，将自由文本描述映射到地学知识库（GCMD、GeoNames、GEE 数据目录与功能词典），使不同脚本中含义相同、表述不同的描述统一为标准化实体。</p> <p>在从 295,943 个脚本中选取的 237 个脚本上，相较 Llama-3、GPT-3.5 和 GPT-4o 等大语言模型基线，实体与关系识别准确率分别最多提升 31.9% 和 12.0%。</p> <p>页首为 Geo-CLASS 框架总览图。图片引自 Zhao et al. (2026), <em>International Journal of Geographical Information Science</em>, 40(3), 808–841, 。</p>https://ankky-zhao.github.io/zh/publications/gee-ops/Sun, 01 Jun 2025 00:00:00 +0000https://ankky-zhao.github.io/zh/publications/gee-ops/<p>通过语法树解析与频繁模式挖掘，从 Earth Engine 脚本中提取算子调用关系，构建算子知识库以提升大语言模型的代码生成效果。</p> <p><strong>引用</strong><br> Hou S., Liang J., <strong>Zhao A.</strong>, Wu H. (2025). GEE-OPs: an operator knowledge base for geospatial code generation on Google Earth Engine powered by large language models. Geo-spatial Information Science. </p> <p><em>封面图：GEE-OPs 算子知识库构建框架。引自 Hou et al. (2025), Geo-spatial Information Science。</em></p>https://ankky-zhao.github.io/zh/publications/geo-fub/Sun, 01 Jun 2025 00:00:00 +0000https://ankky-zhao.github.io/zh/publications/geo-fub/<p>从 Earth Engine 脚本中构建算子-功能知识库，将算子与其承担的功能关联起来，作为大语言模型生成地学代码时的检索上下文。</p> <p><strong>引用</strong><br> Hou S., <strong>Zhao A.</strong>, Liang J., Shen Z., Wu H. (2025). Geo-FuB: constructing an operator-function knowledge base for geospatial code generation with large language models. Knowledge-Based Systems, 319, 113624. </p> <p><em>封面图：Geo-FuB 框架构建流程。引自 Hou et al. (2025), Knowledge-Based Systems。</em></p>https://ankky-zhao.github.io/zh/publications/geocode-gpt/Sun, 01 Jun 2025 00:00:00 +0000https://ankky-zhao.github.io/zh/publications/geocode-gpt/<p>利用专门构建的地理空间代码语料（预训练、指令微调与评测集）微调开源大语言模型，得到专用于地理空间代码生成的模型。</p> <p><strong>引用</strong><br> Hou S., Shen Z., <strong>Zhao A.</strong>, Liang J., Gui Z., Guan X., Li R., Wu H. (2025). GeoCode-GPT: a large language model for geospatial code generation. International Journal of Applied Earth Observation and Geoinformation, 104456. </p> <p><em>封面图：GeoCode-GPT 技术流程。引自 Hou et al. (2025), International Journal of Applied Earth Observation and Geoinformation。</em></p>https://ankky-zhao.github.io/zh/publications/geocogent/Sun, 01 Jun 2025 00:00:00 +0000https://ankky-zhao.github.io/zh/publications/geocogent/<p>一个基于大语言模型的智能体，通过任务规划、地理空间知识检索、代码生成与迭代调试完成地理空间编程任务。</p> <p><strong>引用</strong><br> Hou S., Jiao H., Liang J., Shen Z., <strong>Zhao A.</strong>, Wu H. (2025). GeoCogent: an LLM-based agent for geospatial code generation. International Journal of Geographical Information Science. </p> <p><em>封面图：GeoCogent 总体架构。引自 Hou et al. (2025), International Journal of Geographical Information Science。</em></p>https://ankky-zhao.github.io/zh/publications/geocolab/Sun, 01 Jun 2025 00:00:00 +0000https://ankky-zhao.github.io/zh/publications/geocolab/<p>在协作流程中协调多个分工明确的大语言模型智能体，对地理空间代码进行任务分解、编写、审查与优化。</p> <p><strong>引用</strong><br> Wu H., Jiao H., Hou S., Liang J., Shen Z., <strong>Zhao A.</strong>, et al. (2025). GeoColab: an LLM-based multi-agent collaborative framework for geospatial code generation. International Journal of Digital Earth, 18(2). </p> <p><em>封面图：GeoColab 多智能体系统概念图。引自 Wu et al. (2025), International Journal of Digital Earth。</em></p>https://ankky-zhao.github.io/zh/publications/geo-modeling-kg/Sun, 01 Jun 2025 00:00:00 +0000https://ankky-zhao.github.io/zh/publications/geo-modeling-kg/<p>设计一种由大语言模型抽取实体与关系的地学建模知识图谱，支持建模知识的语义检索与推荐。</p> <p><strong>引用</strong><br> Liang J., Hou S., <strong>Zhao A.</strong>, Xu Q., Xiang L., Li R., Wu H. (2025). Design and application of a semantic-driven geospatial modeling knowledge graph based on large language models. Geo-spatial Information Science. </p> <p><em>封面图：协同增强框架总览。引自 Liang et al. (2025), Geo-spatial Information Science。</em></p>https://ankky-zhao.github.io/zh/publications/gpt-gee-knowledge/Sat, 01 Jun 2024 00:00:00 +0000https://ankky-zhao.github.io/zh/publications/gpt-gee-knowledge/<p>一种 GPT 辅助的流程，从 Google Earth Engine 脚本中抽取地理分析模型的结构化知识，并复用这些知识支持模型构建。</p> <p><strong>引用</strong><br> Liang J., <strong>Zhao A.</strong>, Hou S., Wu H., Jin F. (2024). A GPT-enhanced framework on knowledge extraction and reuse for geographic analysis models in Google Earth Engine. International Journal of Digital Earth, 17(1). </p> <p><em>封面图：提示词设计及与 ChatGPT 的交互。引自 Liang et al. (2024), International Journal of Digital Earth。</em></p>