About
Haitao Yuan (袁海涛 in Chinese) is currently a research fellow working with Prof. Gao Cong at College of Computing and Data Science, NTU. Prior to that, he received his Ph.D. degree in computer science from Tsinghua University, supervised by Prof. Guoliang Li and Prof. Ling Feng. Furthermore, Haitao is fortunate to enjoy a close academic collaboration with Prof. Zhifeng Bao.
Haitao’s research focuses on developing innovative database and machine learning technologies that can effectively and efficiently utilize Spatio-temporal + X (e.g., table, text, and image) data to benefit people in areas such as transportation, healthcare, education, and more. To achieve this goal, Haitao concentrates on three key research fields (SIP):
- Building Scalable multi-modal data management and retrieval systems
- Creating Intelligent multi-modal data manipulation and preparation pipelines
- Developing Practical spatio-temporal models and algorithms for real-world applications
He has published 40+ papers in the top DB/DM conferences and journals (SIGMOD, VLDB, ICDE, KDD, WWW, TKDE, CIKM, etc).
Email: pre_name @163.com (pre_name is yhaitao45)
Office: N4 #B3a-02, 50 Nanyang Avenue, Singapore 639798.
Research Interests
- Spatio-temporal Data Preparation and Management (S&I): trajectory search and join — SIGMOD’26 (a), ICDE’25 (a), ICDE’25 (c), ICDE’19; trajectory generation — VLDB’23 (b), TMC’23; road network generation — VLDB’23 (a); traffic imputation — VLDB’24; data imputation — ICDE’25 (b); map matching — ICDE’26, SIGMOD’25 (a); point cloud data management — SIGMOD’25 (b); geospatial knowledge graph completion — SIGMOD’26 (b).
- AI/LLM-powered Database Optimization (S&P): materialized view advisor — TKDE’22, ICDE’20; query rewrite — VLDB’25; index advisor.
- Spatio-temporal Related/Interdisciplinary Application (I&P): AI4Transportation (e.g., travel time estimation — SIGMOD’25, VLDB’23, CIKM’22, SIGMOD’20; traffic prediction — KDD’25, CIKM’24, DASFAA’24, ICDE’21, DSE; POI recommendation — KDD’26, ICDE’24), AI4Healthcare (e.g., medical diagnosis, molecular prediction) and AI4Education (e.g., mathematical exercise solver — DASFAA’19, SIGMOD’19; knowledge tracing — TOIS’26, TOIS’24).
- Retrieval-Augmented Generation (S&P): cost-aware RAG — ICDE’26; multi-modal vector database.
Selected Awards
- 2025 ACM SIGSPATIAL China “Spatial Intelligent” Rising Star Award (3/10)
- 2021 ACM SIGMOD China Doctoral Dissertation Award (1/2)
- 2021 Best Ph.D Thesis of Tsinghua CS
- 2021 Outstanding Graduate of Beijing
- 2020 National Scholarship
- 2020 Innovative Future Scholarship of Tsinghua CS
- 2019 VMware Scholarship Award
- 2019 Best Paper Award of DASFAA2019
Program Committee Member
- 2026: VLDB, ICDE, KDD, ICML, ICLR, SIGSPATIAL
- 2025: VLDB, KDD, ICML, ICLR, NuerIPS
- 2025: ICDE, KDD
Journal Reviewer
- IEEE Transactions on Knowledge and Data Engineering
- IEEE Transactions on Mobile Computing
Invited Talks
I am happy to give a talk if you are interested in my work. 😊
GoodTP: An Effective Data Selection Framework for Enhancing Trajectory Similarity Learning via Monte Carlo Tree Search. SIGMOD’26, 2026. 06 [Slides]
GeoKGM: A Multimodal Large Language Model for Zero-Shot Knowledge Graph Completion in Geospatial Databases. SIGMOD’26, 2026. 06 [Slides]
Intelligent Management and Analysis of Spatio-temporal Big Data in the Age of Generative AI (In Chinese). SpatialDI’25, 2025. 04 [Slides]
Nuhuo: An Effective Estimation Model for Traffic Speed Histogram Imputation on A Road Network. PVLDB’24, 2024. 08 [Slides]
Route Travel Time Estimation on A Road Network Revisited: Heterogeneity, Proximity, Periodicity and Dynamicity. PVLDB’23, 2023. 09 [Slides]
Automatic Road Extraction with Multi-Source Data Revisited: Completeness, Smoothness and Discrimination. PVLDB’23, 2023. 09 [Slides]
Deep Learnig for ETA Application on Baidu Map. Beijing University of Posts and Telecommunications. 2022. 05.
Big Spatio-temporal Trajectory Data Management and Mining (In Chinese). Renmin University of China, 2022. 01
