1st Large Language Models for Spatial-rich Data Management (LLM+Spatial)

Abstract: Large Language Models are replacing many components in data systems. Spatial data systems are no exception. Important questions arise: Do LLMs replace spatial data systems in their entireties or do we still need spatial data systems? How do we adapt spatial data systems to leverage LLMs and gain the best of both worlds? What new systems requirements do LLMs and ML techniques impose on spatial data systems? This talk addresses these questions, and highlights potential directions for the interplay of spatial data systems and LLMs to achieve a win-win scenario (1+1=3).

Bio: Walid G. Aref is a professor of computer science at Purdue. His research interests are in extending the functionality of database systems in support of emerging applications, e.g., spatial, spatio-temporal, graph, and sensor databases. He is also interested in query processing, indexing, data streaming, and geographic information systems (GIS). Walid's research has been supported by the National Science Foundation, the National Institute of Health, Purdue Research Foundation, CERIAS, Panasonic, and Microsoft Corp. In 2001, he received the CAREER Award from the National Science Foundation and in 2004, he received a Purdue University Faculty Scholar award. Walid is a member of Purdue's CERIAS. He has served as Editor-in-Chief of the ACM Transactions of Spatial Algorithms and Systems (ACM TSAS), an editorial board member of the Journal of Spatial Information Science (JOSIS), and an editor of the VLDB Journal and the ACM Transactions of Database Systems (ACM TODS). Walid has won several best paper awards including the 2016 VLDB ten-year best paper award. He is a Fellow of the IEEE, and a member of the ACM. Between 2011 and 2014, Walid has served as the chair of the ACM Special Interest Group on Spatial Information (SIGSPATIAL).

Abstract: Maps are powerful, but making sense of them has traditionally required specialized expertise in GIS software, complex query languages, and significant manual effort. Advances in large language models (LLMs) and generative AI are beginning to change this dynamic, opening new ways of working with spatial data that are far more intuitive. Instead of relying on specialized tools, users can now describe the data they need or write complex geographic questions, and intelligent systems can translate those intentions into concrete results. This talk will highlight recent progress in three key directions: generating realistic spatial datasets from textual descriptions, answering complex questions that combine spatial reasoning with external knowledge, and automatically creating styles that make map visualizations more effective and interpretable. Taken together, these advances illustrate a new paradigm where geospatial data can be explored and understood through a natural and accessible interface.

Bio: Ahmed Eldawy is an Associate Professor in Computer Science at the University of California Riverside. His research interests lie in the broad area of databases with a focus on big data management and spatial data processing. Ahmed led the research and development in many open source projects for big spatial data exploration and visualization including UCR-Star, an interactive repository for geospatial data with nearly four terabytes of publicly available data. He is a recipient of the highly prestigious NSF CAREER award, the 10-year Influential Paper Award in ICDE 2025, and the Best Demo award in SIGSPATIAL 2020. His work is supported by the National Science Foundation (NSF) and the US Department of Agriculture (USDA).

Abstract: The talk will cover the following research problems on geospatial entity representation: 1) Geospatial Entity Representation for point objects, trajectory, and regions and their applications, e.g., spatial keyword search, POI recommendation, speed inference, region population estimations, etc. 2) Foundation Models for Geospatial Applications and Efforts toward City Foundation Models. The first part primarily concentrates on learning representations to facilitate geospatial entity querying and analysis. The second part focuses on self-supervised learning approaches applied to geospatial entities, and several research attempts towards city foundation models.

Bio: Gao Cong is currently a Professor in the College of Computing and Data Science (CCDS) at Nanyang Technological University (NTU). He serves as the head of Division of Data Science, CCDS, NTU. He previously worked at Aalborg University, Denmark, Microsoft Research Asia, and the University of Edinburgh. His current research interests include AI4DB, spatial data management, spatial-temporal data mining, and recommendation systems. His citation in Google Scholar was over 20,000 with H-index 75. He received SIGIR'25 test of time award honourable mention award, and the best paper runner-up awards at the WSDM'20 and WSDM'22 conferences for two of his research papers. He served as a PC co-chair for ICDE'2022, the associate general chair of KDD'21, a PC co-chair for E&A track of VLDB 2014, and a PC vice-Chair for ICDE'18. He serves as an associate editor for ACM Transactions on Database Systems (TODS) and IEEE TKDE.