Welcome! I'm Jiyoon Pyo

Ph.D. Student at University of Minnesota-Twin Cities

I am a third-year Ph.D. student in Computer Science at the University of Minnesota–Twin Cities, where I am a member of the Knowledge Computing Lab advised by Prof. Yao-Yi Chiang.

My research focuses on advancing the understanding of how large language models (LLMs) and large vision-language models (LVLMs) process geospatial information. I am especially interested in the capabilities and limitations of these models in performing geospatial reasoning tasks, such as interpreting topological relationships, recognizing spatial features, and engaging in multi-hop reasoning over map content. My current work involves developing benchmarks and evaluation methods to assess LVLM map-reasoning abilities, and exploring techniques to enhance their spatial understanding and reliability on domain-specific tasks. Beyond my current work, I am also interested about linguistic variations, including regional terminology differences and cultural influences on data and model interpretation.

I earned my B.Sc. in Computer Science from the University of Minnesota and both my B.Eng and M.Eng in Electrical Engineering from The Cooper Union. My Master's thesis, Detection and Replacement of Neologisms for Translation, supervised by Dr. Carl Sable, explored methods for identifying newly coined terms to improve machine translation quality. Additionally, I worked as a research assistant in the Language and Voice Lab at Reykjavik University with Dr. Stefán Ólafsson, where I helped evaluate multilingual language models and their effectiveness for lower-resourced languages.

News

đź“„ Our paper 'Augmenting Human-Centered Racial Covenant Detection and Georeferencing with Plug-and-Play NLP Pipelines' is accepted to GeoHCC

📄 Our paper “Exploiting LLMs and Semantic Technologies to Build a Knowledge Graph of Historical Mining Data” is accepted to ISWC 2025

🎓 I completed my Master’s in Computer Science at the University of Minnesota

🎓 I passed my Ph.D. qualification exam and am now officially a Ph.D. candidate

📄 Our paper “Leveraging Large Language Models for Generating Labeled Mineral Site Record Linkage Data” is accepted to GeoAI 2024