PhD in Cartography
2025–present
Technical University of Munich
Supervisor Prof. Liqiu Meng · Mentor Prof. Yu Feng
M.Sc. in Cartography and GIS
2022–2025
Wuhan University
Supervisors Prof. Zhipeng Gui and Prof. Huayi Wu
B.Eng. in Geodesy and Geomatics Engineering
2018–2022
Wuhan University
Top 5% of the cohort by GPA
Peer-reviewed papers, with figures and code where available
Google Scholar profile
Zhao A., Gui Z., Liang J., Liu Y., Peng D., Wei W., Hou S., Wu H.
A pipeline that coordinates large and small language models to extract reusable geoprocessing modelling knowledge from crowdsourced Google Earth Engine scripts.
Gui Z., Huang Z., Chen H., Peng D., Liu Y., Fang G., Lan Q., Zhao A., Wu H.
Parallelises the clustering-by-local-direction-centrality algorithm on Apache Spark with spatial partitioning and performance optimisations for large datasets.
Wei W., Gui Z., Wu C., Zhao A., Peng D., Wu H.
Learns several dynamic visual sub-embeddings per image to capture varied semantics and adds a fast re-ranking stage to improve image-text retrieval.
Hou S., Liang J., Zhao A., Wu H.
Mines operator call relationships from Earth Engine scripts with syntax-tree parsing and frequent-pattern analysis, forming a knowledge base that improves LLM code generation.
Hou S., Zhao A., Liang J., Shen Z., Wu H.
Builds an operator-function knowledge base from Earth Engine scripts, linking operators to the functions they serve, and uses it as retrieval context for geospatial code generation with LLMs.
Hou S., Shen Z., Zhao A., Liang J., Gui Z., Guan X., Li R., Wu H.
Fine-tunes an open LLM on purpose-built geospatial code corpora for pretraining, instruction tuning and evaluation, giving a model specialised in generating geospatial code.
Hou S., Jiao H., Liang J., Shen Z., Zhao A., Wu H.
An LLM-based agent that plans, retrieves geospatial knowledge, and generates and iteratively debugs code to complete geospatial programming tasks.
Wu H., Jiao H., Hou S., Liang J., Shen Z., Zhao A., et al.
Coordinates role-specialised LLM agents in a collaborative workflow that decomposes, writes, reviews and refines geospatial code.
Liang J., Hou S., Zhao A., Xu Q., Xiang L., Li R., Wu H.
Designs a geospatial modeling knowledge graph whose entities and relations are extracted with LLMs, supporting semantic retrieval and recommendation of modeling knowledge.
Liu Y., Gui Z., Xu Y., Gao S., Zhao A., et al.
Introduces a Trajectory Trait Scale inspired by the five-factor personality model, scoring drivers’ extroversion, openness, neuroticism and conscientiousness from GPS trajectory features.
Liang J., Zhao A., Hou S., Wu H., Jin F.
A GPT-assisted workflow that extracts structured knowledge about geographic analysis models from Google Earth Engine scripts and reuses it to support model construction.
Gui Z., Liu X., Zhao A., et al.
Recognises users’ map retrieval intentions from relevance feedback, with geographic ontologies guiding intention inference and its iterative updates.
Yan L., Yang J., Zhang Y., Zhao A., Li X.
Normalises radiometry across Gaofen sensors by detecting unchanged pixels with change detection and a chi-square test, then fitting a relative normalisation model.
Open-source projects on GitHub
A bilingual learning map that takes readers from first contact with LLM agents to research topics and their own projects.
A bilingual tutorial that goes from the first HTTP request to a complete scraping project, with both hand-written and AI-assisted code.
Oct 2026 · Launched this website. I am open to collaboration on trustworthy AI and LLM agents, see Open to Collaboration.
Oct 2026 · Released the Web Scraping Roadmap, a bilingual guide from first request to a finished scraping project.
Sep 2026 · Released the LLM Agent Roadmap, an open learning map of LLM agents with curated resources and a key-papers timeline.
Nov 2025 · First-author paper on coordinating large and small language models for geoprocessing knowledge extraction published in IJGIS.
Sep 2025 · Started my PhD at the Chair of Cartography and Visual Analytics, TUM.
Supervision · Master’s theses on benchmarking and evaluating explanations of geospatial AI.
Teaching · Topic design for the seminar Mapping for a Sustainable World, TUM. Earlier teaching assistant for the Artificial Intelligence Summer School at the University of Cambridge (2021), the International Geoinformatics Summer School (2022) and Data Structure and Programming in C++ (2024–2025) at Wuhan University.
Notes on turning my own reading list into a public, bilingual roadmap.
I would like to work with people who share an interest in trustworthy and explainable AI, LLM and multimodal agents and newer agent topics such as evaluation, reliability and reasoning, or location privacy in the age of AI. Projects close to GeoAI are welcome, and so are questions from the broader AI side where spatial data is one testbed among others. PhD students, postdocs and industry researchers are all welcome, for a joint paper, a dataset or benchmark, or an open-source tool. If this overlaps with your work, a short email about what you are working on is the best way to start.