Publications [Home]

2026

• FDABench: A Benchmark for Data Agents on Analytical Queries over Heterogeneous Data DS4AI

• ARCADE: A Real-Time Data System for Hybrid and Continuous Query Processing across Diverse Data Modalities DS4AI

• CARROT: A Learned Cost-Constrained Retrieval Optimization System for RAG DS4AI

• MULLER: A Multimodal Data Lake Format for Collaborative AI Data Workflows DS4AI

• ARCADE: A Real-Time Data System for Hybrid and Continuous Query Processing across Diverse Data Modalities DS4AI

• Demonstrating TOFFEE: A Learned System for Synthesizing Data Agent Trajectories at Scale DS4AI

• Are Learned DBMS Components Robust to Workload Drift? AI4DB

• TATA: An Efficient Framework for Task Transfer in Query Plan Representation AI4DB

• Towards Industrial-Scale Parametric Query Optimization [Industry] AI4DB

• SOLAR: Efficient Spatial Queries on Real-time LSM-based Storage SDMAI4DB

• GoodTP: An Effective Data Selection Framework for Enhancing Trajectory Similarity Learning via Monte Carlo Tree Search STDM

• Traj-MLLM: Can Multimodal Large Language Models Reform Trajectory Data Mining STDM

• Cross-city Time Series Forecasting with Retrieval-Augmented Large Language Models STDM

• UrbanMFM: Spatial Graph-based Multiscale Foundation Models for Learning Generalized Urban Representation STDM

• Disentangling Dynamics: Advanced, Scalable and Explainable Imputation for Multivariate Time Series STDM

• CausalPOI: Spatio-Temporal Graph-Based Causal Modeling for Cold-Start POI Check-in Forecasting POIRecSTDM

2025

• DEG: Efficient Hybrid Vector Search Using the Dynamic Edge Navigation Graph DS4AI

• MAST: Towards Efficient Analytical Query Processing on Point Cloud Data DS4AI SDM

• Demonstrating MAST: An Efficient System for Point Cloud Data Analytics DS4AI

• CAMAL: Optimizing LSM-trees via Active Learning AI4DB

• NEXT: A New Secondary Index Framework for LSM-based Data Storage DS4AI

• LLM-R2: A Large Language Model Enhanced Rule-based Rewrite System for Boosting Query Efficiency AI4DB

• RankPQO: Learning-to-Rank for Parametric Query Optimization AI4DB

• BT-Tree: A Reinforcement Learning Based Index for Big Trajectory Data STDMAI4DB

• GeoBloom: Revisiting Lightweight Models for Geographic Information Retrieval SDM

• LIST: learning to index spatio-textual data for embedding based spatial keyword queries SDM

• Reconciling Geospatial Prediction and Retrieval via Sparse Representations SDMSTDM

• Self-supervised representation learning for geospatial objects: A survey SDMSTDM

• Semantic-Enhanced Representation Learning for Road Networks With Temporal Dynamics SDMSTDM

• ST-LLM+: Graph Enhanced Spatio-Temporal Large Language Models for Traffic Prediction STDM

• Enhancing Large Language Models for Mobility Analytics with Semantic Location Tokenization STDM

• Foundation Models for Spatio-Temporal Data Science: A Tutorial and Survey STDM

• PLMTrajRec: A Scalable and Generalizable Trajectory Recovery Method with Pre-trained Language Models STDM

• SMARTraj: A Stable Multi-City Adaptive Method for Multi-View Spatio-Temporal Trajectory Representation Learning STDM

• TransferTraj: A Vehicle Trajectory Learning Model for Region and Task Transferability STDM

• FSTLLM: Spatio-Temporal LLM for Few Shot Time Series Forecasting STDM

• Exploring Progress in Multivariate Time Series Forecasting: Comprehensive Benchmarking and Heterogeneity Analysis STDM

• ARTS: A General and Efficient Multi-Task Self-Prompt Framework for Explainable Sequential Recommendation POIRec

• HubGT: Fast Graph Transformer with Decoupled Hierarchy Labeling DMSA

2024

• Lemo: A Cache-Enhanced Learned Optimizer for Concurrent Queries AI4DB

• PLATON: Top-down R-tree Packing with Learned Partition Policy AI4DB

• Selectivity Estimation for Queries Containing Predicates over Set-Valued Attributes AI4DB

• A Comparative Study and Component Analysis of Query Plan Representation Techniques in ML4DB Studies AI4DB

• Nuhuo: An Effective Estimation Model for Traffic Speed Histogram Imputation on A Road Network SDMSTDM

• Road Network Representation Learning with the Third Law of Geography SDMSTDM

• Zero-shot urban function inference with street view images through prompting a pretrained vision-language model SDMSTDM

• City Foundation Models for Learning General Purpose Representations from OpenStreetMap STDM

• On the opportunities and challenges of foundation models for geoai (vision paper) STDM

• TERI: An Effective Framework for Trajectory Recovery with Irregular Time Intervals STDM

• Trajectory Similarity Measurement: An Efficiency Perspective STDM

• AdapTraj: A Multi-Source Domain Generalization Framework for Multi-Agent Trajectory Prediction STDM

• Collectively Simplifying Trajectories in a Database: A Query Accuracy Driven Approach STDM

• SAGDFN: A Scalable Adaptive Graph Diffusion Forecasting Network for Multivariate Time Series Forecasting STDM

• AirPhyNet: Harnessing Physics-Guided Neural Networks for Air Quality Prediction STDM

• MMPOI: A Multi-Modal Content-Aware Framework for POI Recommendations POIRec

• UnifiedSSR: A Unified Framework of Sequential Search and Recommendation POIRec

• On Evaluation Metrics for Diversity-enhanced Recommendations POIRec

• Quantum Algorithm for Maximum K-Plex Problem DMSA

2023

• The RLR-Tree: A Reinforcement Learning Based R-Tree for Spatial Data AI4DBSDM

• WISK: A Workload-aware Learned Index for Spatial Keyword Queries AI4DBSDM

• Learned Index Benefits: Machine Learning Based Index Performance Estimation AI4DB

• Region Embedding with Intra and Inter-View Contrastive Learning. AI4DB

• Effectiveness Perspectives and a Deep Relevance Model for Spatial Keyword Queries SDM

• Online Anomalous Subtrajectory Detection on Road Networks with Deep Reinforcement Learning SDMSTDM

• Urban Region Representation Learning with OpenStreetMap Building Footprints SDMSTDM

• Multivariate Time-series Imputation with Disentangled Temporal Representations STDM

• Mining Geospatial Relationships from Text STDM SDM

• Towards Designing and Learning Piecewise Space-Filling Curves AI4DB SDM

2022

• SAM: Database Generation from Query Workload with Supervised Autoregressive Model AI4DB

• Cardinality Estimation in DBMS: A Comprehensive Benchmark Evaluation AI4DB

• QueryFormer: A Tree Transformer Model for Query Plan Representation AI4DB

• Unsupervised Selectivity Estimation by Integrating Gaussian Mixture Models and Autoregressive Model AI4DB

• Example-based Spatial Pattern Matching SDM

• Geospatial Entity Resolution SDMDMSA

• Similar Trajectory Search with Spatio-temporal Deep Representation Learning STDM

• Points-of-Interest Relationship Inference with Spatial-enriched Graph Neural Networks POIRec

• The Datasets Dilemma: How Much Do We Really Know About Recommendation Datasets? POIRec

• DMCS : Density Modularity based Community Search DMSA

• Entity Resolution with Hierarchical Graph Attention Networks DMSA

• ABC : Attributed Bipartite Co-clustering DMSA

2021

• A Unified Deep Model of Learning from both Data and Queries for Cardinality Estimation AI4DB

• Location- and keyword-based querying of geo-textual data: a survey SDM

• Robust Road Network Representation Learning: When Traffic Patterns Meet Traveling Semantics SDMSTDM

• STAR: A Cache-based Distributed Warehouse System for Spatial Data Streams SDM

• Trajectory Simplification with Reinforcement Learning STDM

• Error-Bounded Online Trajectory Simplification with Multi-Agent Reinforcement Learning STDM

• A Linear Time Approach to Computing Time Series Similarity based on Deep Metric Learning STDM

• Learning Dynamics and Heterogeneity of Spatial-Temporal Graph Data for Traffic Forecasting STDM

• A Survey on Trajectory Data Management, Analytics, and Learning STDM

• Sinkhorn Collaborative Filtering POIRec

2020

• STAR: A Distributed Stream Warehouse System for Spatial Data (Demo) SDM

• SSTD: A Distributed System on Streaming Spatio-Textual Data SDM

• Spatial Transition Learning on Road Networks with Deep Probabilistic Models SDMSTDM

• Efficient and Effective Similar Subtrajectory Search with Deep Reinforcement Learning STDM

• Online Anomalous Trajectory Detection with Deep Generative Sequence Modeling STDM

• Context-aware Deep Model for Joint Mobility and Time Prediction STDM

• PGeoTopic: A Distributed Solution for Mining Geographical Topic Models STDMDMSA

• HyperML: A Boosting Metric Learning Approach in Hyperbolic Space for Recommender Systems POIRec

• Global Context Enhanced Graph Nerual Networks for Session-based Recommendation POIRec

• HME: A Hyperbolic Metric Embedding Approach for Next-POI Recommendation POIRec

• Densely Connected User Community and Location Cluster Search in Location-Based Social Networks DMSA

• Structural Relationship Representation Learning with Graph Embedding for Personalized Product Search DMSA

• Deterministic Inference of Topic Models via Maximal Latent State Replication DMSA

2019

• Finding attribute-aware similar regions for data analysis SDM

• SURGE: Continuous Detection of Bursty Regions Over a Stream of Spatial Objects SDM

• Exploring Market Competition over Topics in Spatio-Temporal Document Collections STDMDMSA

• Computing Trajectory Similarity in Linear Time: A Generic Seed-Guided Neural Metric Learning Approach STDM

• Learning Travel Time Distributions with Deep Generative Model STDM

• Interact and Decide: Medley of Sub-Attention Networks for Effective Group Recommendation POIRec

• Effective and Efficient Sports Play Retrieval with Deep Representation Learning DMSA

2018

• Efficient Selection of Geospatial Data on maps for Interactive Visualized Exploration SDM

• On Spatial Pattern Matching SDM

• SURGE: Continuous Detection of Bursty Regions over a Stream of Spatial Objects SDM

• Efficient Similar Region Search with Deep Metric Learning SDM

• Linking Fine-Grained Locations in User Comments SDMDMSA

• Deep Representation Learning for Trajectory Similarity Computation [pdf] (code) STDM

• Periodic-CRN: A Convolutional Recurrent Model for Crowd Density Prediction with Recurring Periodic Patterns STDMDMSA

• Reverse k Nearest Neighbor Search over Trajectories STDM

• ANR: Aspect-based Neural Recommender (code ) POIRec

• Finding Seeds and Relevant Tags Jointly: For Targeted Influence Maximization in Social Networks DMSA

• PANDA: A System for Partial Topology-based Search on Large Networks (Demo) DMSA

• Inf2vec: Latent Representation Model for Social Influence Embedding DMSA

• High-dimensional Similarity Learning via Dual-sparse Random Projection DMSA

2017

• Distributed Publish/Subscribe Query Processing on the Spatio-Textual Data Stream SDM

• Discovering Fine Grained Pollution Sources and Propagation Patterns in Urban Area STDMDMSA

• PRED: Periodic Region Detection for Mobility Modeling of Social Media Users STDMDMSA

• Influence maximization in trajectory databases STDMDMSA

• An Experimental Evaluation of Point-of-interest Recommendation in Location-based Social Networks (code and dataset) POIRecDMSA

• A General Model for Out-of-town Region Recommendation POIRec

• POI2Vec: Geographical Latent Representation for Predicting Future Visitors POIRec

• Distributed Algorithms on Exact Personalized PageRank DMSA

• PANDA: Towards Partial Topology-based Search on Large Networks in a Single Machine DMSA

• Analyzing Sentiments in One Go: A Supervised Joint Topic Modeling Approach DMSA

• Biclustering: An application of Dual Topic Models DMSA

2016

• Towards Best Region Search for Data Exploration SDM

• Tutorial: Querying Geo-Textual Data: Spatial Keyword Queries and Beyond (Tutorial slides ) SDM

• A System for Region Search and Exploration (Demo website) SDM

• Towards Personalized Maps: Mining User Preferences from Geo-textual Data (Demo website) SDM

• Recognition and Linking of Fine-Grained Locations from Tweets [pdf] SDMDMSA

• Annotating Points of Interest with Geo-tagged Tweets SDMSTDM

• Efficient Processing of Location-Aware Group Preference Queries SDM

• Invited paper: Querying and Mining Geo-textual Data for Exploration: Challenges and Opportunities SDM

• Topic Exploration in Spatio-Temporal Document Collections STDMDMSA

• A General Recommendation Model for Heterogeneous Networks POIRec

• Learning to Find Topic Experts in Twitter via Different Relations DMSA

• ConTrack: A Scalable Method For Tracking Multiple Concepts In Large Scale Multidimensional Data DMSA

2015

• Efficient Algorithms for Answering the m-Closest Keywords Query [pdf] SDM

• Temporal Spatial-Keyword Top-k Publish/Subscribe [pdf] (Datasets) SDM

• Efficient Processing of Spatial Group Keyword Queries SDM

• SAR: A Sentiment-Aspect-Region Model for User Preference Analysis in Geo-tagged Reviews [pdf] STDMDMSA

• On Information Coverage for Location Category Based Point-of-Interest Recommendation POIRec

• A General Graph-based Model for Recommendation in Event-based Social Networks [pdf] (Datasets ) POIRecDMSA

• Rank-GeoFM: A Ranking based Geographical Factorization Method for Point of Interest Recommendation [pdf] POIRec

• Where you Instagram? Associating Your Instagram Photos with Points of Interest [pdf] POIRec

• Personalized Ranking Metric Embedding for Next New POI Recommendation [pdf] POIRec

• Who, where, when and what: a nonparametric bayesian approach to context-aware recommendation and search for twitter users. [pdf] (Dataset) POIRecDMSA

• Diversity-Aware Top-k Publish/Subscribe for Text Stream [pdf] DMSA

• A Tri-Role Topic Model for Domain-Specific Question Answering DMSA

• DynaDiffuse: A Dynamic Diffusion Model for Continuous Time Constrained Influence Maximization DMSA

• Mining User Intents in Twitter: A Semi-Supervised Approach to Inferring Intent Categories for Tweets [pdf] DMSA

2014

• Retrieving Regions of Interest for User Exploration SDM

• Distributed Spatial Keyword Querying on Road Networks SDMSTDM

• Efficient algorithms and cost models for reverse spatial-keyword k-nearest neighbor search SDM

• SOPS: A System for Efficient Processing of Spatial-Keyword Publish/Subscribe SDM

• COM: A Generative Model for Group Recommendation POIRec

• Graph-based Point-of-interest Recommendation with Geographical and Temporal Influences (datasets) POIRecDMSA

• In Search of Influential Event Organizers in Online Social Networks DMSA

• Coarse-to-Fine Review Selection via Supervised Joint Aspect and Sentiment Model DMSA

• Tagging Your Tweets: A Probabilistic Modeling of Hashtag Annotation in Twitter DMSA

• Influence Maximization with Novelty Decay in Social Networks DMSA

• Where your photo is taken: Geolocation prediction for social images DMSA

2013

• An efficient query indexing mechanism for filtering geo-textual data SDM

• Spatial Keyword Query Processing: An Experimental Evaluation [PDF] (code) SDM

• KORS: Keyword-aware Optimal Route Search System SDM

• Efficient proximity detection among mobile objects in road networks with self-adjustment methods SDMSTDM

• Who, where, when and what: discover spatio-temporal topics for twitter users (datasets) STDMDMSA

• Time-aware point-of-interest recommendation (datasets) POIRec

• Influence Spreading Path and its Application to the Time Constrained Social Influence Maximization Problem and Beyond (Source code ) DMSA

2012

• Keyword-aware Optimal Route Search [PDF] SDM

• Shortest Path and Distance Queries on Road Networks: An Experimental Evaluation. SDMSTDM

• A framework for efficient spatial web object retrieval SDM

• DISKs: A System for Distributed Spatial Group Keyword Search on Road Networks SDMSTDM

• SWORS: A System for Efficient Retrieval of Relevant Spatial Web Objects [PDF] SDM

• Author Name Disambiguation Using a New Categorical Distribution Similarity [PDF] DMSA

• Overlapping Decomposition for Causal Graphical Modeling DMSA

• Enhancing naive bayes with various smoothing methods for short text classification [PDF] DMSA

• Category Hierarchy Maintenance: A Data-Driven Approach [PDF] Annotated dataset here DMSA

• Will This #Hashtag Be Popular Tomorrow? DMSA

• One seed to find them all: Mining Opinion Features via Association [PDF] DMSA

• Topic-Driven Reader Comments Summarization DMSA

• Approaches to Exploring Category Information for Question Retrieval inCommunity Question-Answer Archives [PDF] DMSA

• Functional Feature Extraction and Chemical Retrieval DMSA

• 3D Subspace Clustering in Value Investing DMSA

• An Evaluation of Classification Models for Question Topic Categorization [PDF] DMSA

• Time Constrained Influence Maximization in Social Networks [PDF] ( The algorithm in the ICDM paper is extended to find influential nodes for the traditional influence maximization problem without time constraints, which appears in our TKDE paper. Source code ) DMSA

2011

• On the Complexity of View Update Analysis and its Application to Annotation Propagation AI4DB

• Partial Evaluation for Distributed XPath Query Processing and Beyond AI4DB

• Collective Spatial Keyword Querying. [PDF] (Source code here ) SDM

• Efficient Continuously Moving Top-K Spatial Keyword Query Processing SDM

• Joint Top-K Spatial Keyword Query Processing, SDM

• Reverse Spatial and Textual k Nearest Neighbor Search [PDF] DMSA

• Integrating Community Question and Answer Archives DMSA

• Simulated Annealing based Influence Maximization in Social Networks DMSA

• A Survey on Enhanced Subspace Clustering DMSA

• Centroid-based Actionable 3D Subspace Clustering DMSA

2010

• Retrieving Top-k Prestige-Based Relevant Spatial Web Objects SDM

• Mining Significant Semantic Locations From GPS Data, DMSA

• Community-based Greedy Algorithm for Mining Top-K Influential Nodes in Mobile Social Networks [PDF] DMSA

• A Generalized Framework of Exploring Category Information for Question Retrieval in Community Question Answer Archives. (supplementary materials on dataset) DMSA

• Content-enriched Classifier for Web Video Classification DMSA

• Exploring Domain-specific Term Weight in Archived Question Search. DMSA

• Evolutionary Taxonomy Construction from Dynamic Tag Space DMSA

• ISIS: A New Approach for Efficient Similarity Search in Sparse Databases DMSA

• Using Transactional Data from ERP Systems for Expert Finding [PDF] DMSA

2009

• A Revisit of Query Expansion with Different Semantic Levels AI4DB

• Efficient Retrieval of the Top-k Most Relevant Spatial Web Objects SDM

• The Use of Categorization Information in Language Models for Question Retrieval. (supplementary materials on dataset) DMSA

• Routing Questions to Right Users in Online Communities DMSA

• Efficient Algorithms for Computing Link-based Similarity in Real World Networks. DMSA

2008

• Updating Recursive XML Views of Relations. AI4DB

• Semantic Similarity Based on Compact Concept Ontology. DMSA

• Finding Question-Answer Pairs from Online Forums DMSA

• Using Conditional Random Fields to Extract Contexts and Answers of Questions from Online Forums. (supplementary materials on data) DMSA

2007

• Distributed Query Evaluation with Performance Guarantees. AI4DB

• Query XML with Update Syntax. AI4DB

• Updating recursive XML views of relations. AI4DB

• Query and Update Through XML Views. AI4DB

• Improving Data Quality: Consistency and Accuracy. DMSA

• Detecting Erroneous Sentences using labeled sequential Patterns and Tree Patterns. DMSA

• Detecting Erroneous Sentences using Automatically Mined Sequential Patterns. DMSA

2006

• Using Partial Evaluation in Distributed Query Evaluation. AI4DB

• An Estimation System for XPath Expressions. AI4DB

• Annotation Propagation Revisited for Key Preserving Views. AI4DB

• Composite Acoustic Features for Efficient Music Similarity Query. AI4DB

• Summarizing frequent patterns using profiles. DMSA

2005

• Mining Top-k Covering Rule Groups for Gene Expression Data. DMSA

• On effective E-mail Classification via Neural Networks. DMSA

2004

• FARMER: Fining Interesting Association Rule Groups by Row Enumeration in Biological Datasets. DMSA

• Semantic Mining and Analysis of Gene Expression Data (Demo). DMSA

• Go Green: Recycle and Reuse Frequent Patterns. DMSA

• Large Incremental Maintenance of Quotient Cube for Sum and Median. DMSA

• COBBLER: Combining Column and Row Enumeration for Closed Pattern Discovery. DMSA

• Mining Frequent Closed Patterns in Microarray Data. DMSA

• Semi-Supervised Text Classification Using Partitioned EM. DMSA

2003

• CARPENTER: Finding Closed Patterns in Long Biological Datasets. DMSA

2002

• Discovering frequent substructures from hierarchical semi-structured data. DMSA

• Speed-up Iterative Frequent Itemset Mining with Constraint Changes DMSA