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

Workshop Overview

The importance of spatio-temporal data has increased significantly in various scientific fields, such as climate research, biodiversity, and the social sciences, primarily due to improvements in data collection and accessibility. Despite the opportunities for new scientific insight, researchers often face the challenge of inadequate tools and interfaces for managing, integrating, and analyzing spatio-temporal data. Recently, the emergent abilities of LLMs represent a pivotal point that is to significantly affect the academic and industrial communities. The vast amount of knowledge in spatial-rich data is not used to train and tune LLMs, and, spatio-temporal databases are not able to access and operate on the facts contained in the LLMs. This workshop aims to provide new insight into techniques from spatial-rich data and large language models to improve advances in spatial-rich data management and predictive models.

Organization

Workshop Co-Chairs

Jianqiu Xu

Nanjing University of Aeronautics and Astronatucis, P.R.China

jianqiu@nuaa.edu.cn

Cheng Long

Nanyang Technological University, Singapore

c.long@ntu.edu.sg

Bernhard Seeger

University of Marburg, Germany

seeger@mathematik.uni-marburg.de

Yongxin Tong

Beihang University, P.R.China

yxtong@buaa.edu.cn

Call for Papers

The goal is to advance the understanding of how LLMs and spatial-rich data management can cooperatively contribute to novel data science solutions. Topics of interest include, but are not limited to:

• Spatial-rich Data Foundation Model
• Enhance LLMs by Spatial-rich Data
• Spatial-rich Data Quality, Anomaly Detection, and Imputation with LLMs
• Retrieval-augmented Models for Geospatial Applications
• NL2SQL for Spatio-temporal Data
• Spatial and Temporal-spatial Contextual Reasoning with LLMs
• Embedding Learning for Geospatial Data with LLMs
• Fine-tuning LLMs on Domain-specific Geospatial Data
• Benchmarking of LLMs + Spatio-temporal Databases
• Optimizing Spatio-temporal Databases with LLMs
• Cases Studies and Applications of LLMs + Spatial-rich Data
• Visions for LLMs + Spatio-temporal Databases

Program

Keynotes

Keynote 1

---

Walid G. Aref

Purdue University

Keynote 2

---

Gao Cong

Nanyang Technological University

Program Committee

• Sheng Wang, Wuhan University, P.R.China
• Mahmoud Sakr, Free University of Brussels, Belgium
• Ziqiang Yu, Yantai University, P.R.China
• Man Lung Yiu, Hong Kong Polytechnic University, Hong Kong, P.R.China
• Zheng Wang, Huawei Singapore Research Center, Singapore
• Andreas Züfle, Emory University, U.S.A
• Chenxi Liu, Nanyang Technological University, Singapore
• Qianxiong Xu, Nanyang Technological University, Singapore
• Liang Zhang, HEC Paris, Singapore
• Ziquan Fang, Zhejiang University, P.R.China
• Yuren Mao, Zhejiang University, P.R.China
• Matthias Renz, University of Kiel, Germany
• Amr Magdy, University of California Riverside, U.S.A
• Tianyi Li, Aalborg University, Denmark
• Ahmed Eldawy, University of California Riverside, U.S.A

Submission Instructions

Prospective authors are invited to submit original research papers that address the topics of interest for the workshop. For authors submitting their papers (.pdf format), please format using the style file. We call for two types of papers:

1. Vision Papers (up to 4 pages, plus additional pages for the reference pages)
2. Regular Research Papers (from 6 to 8 pages, plus additional pages for the reference pages)

Submission site: https://cmt3.research.microsoft.com/LLMSpatial2025

Accepted papers will be published in the VLDB Workshop Proceedings. At least one author of each accepted paper is expected to register for VLDB 2025 and present the paper in person.

Review Process

We will enforce a rigorous peer and single-anonymous review process. All manuscripts submitted to our workshop will be reviewed by at least two PC members. Plagiarism Detection Tools will be used to check the content of the submitted manuscripts against previous publications. Papers will be evaluated according to the following aspects:

1. relevance to the workshop topic
2. scientific novelty
3. technical soundness
4. appropriateness and adequacy in terms of literature review and analysis
5. presentation

Handling of Conflicts of Interest

We will follow the conflict of interest policy for ACM publications.

Important Dates (AoE)

• Submission deadline: May 15, 2025
• Notification of acceptance: June 15, 2025
• Camera-ready Version of Accepted Papers: July 1, 2025

Acknowledgement: The Microsoft CMT service was used for managing the peer-reviewing process for this conference. This service was provided for free by Microsoft and they bore all expenses, including costs for Azure cloud services as well as for software development and support.

Accepted Papers

TBA