Research Interests

Zhongle’s research lives at the intersection of data systems and artificial intelligence, where he spends most of his time trying to make AI more efficient, more usable, and a little less expensive to run. He is interested in how learning and inference workloads interact with data and system infrastructures, and in developing techniques that improve model efficiency, data utilization, and overall system performance.

Zhongle takes a data-centric and system-aware view of AI, which means he often looks for efficiency gains in places other than the model itself. Instead of redesigning neural networks, he focuses on how data is organized, accessed, and moved, and how system-level decisions quietly dominate performance. Many of his research questions boil down to a simple theme: how to make AI workloads run faster, cheaper, and more reliably by being smarter about data and systems.

Earlier in his career, Zhongle focused on efficiency and integrity in analytical systems. He later spent nearly four years wandering into startup environments, a memorable period that shaped his perspective on real systems, while gently slowing academic momentum. Along the way, he guided dozens of system and data engineers, reinforcing his belief that good systems must be usable, correct, and efficient in practice.

Selected Publications

Benefiting from a system research perspective, Zhongle maintains sustained collaborations with industry, and often acts closely involved in every paper that carries his name. The marks below indicate research directions and industry collaboration.

Marks: AI Systems & InfrastructureData Systems & AnalyticsHealthcare & Domain AIIndustry Collaboration

Conference

C29J Yuan, et al.. Query as Anchor: Scenario-Adaptive User Representation via Large Language Model. KDD 2026 Industry Collaboration

C28C He, et al.. FOUNDv2: Learning Unified User Quantized Tokenizers for User Representation. KDD 2026 Oral Industry Collaboration

C27C Liu, et al.. DIYHealth Suite: Dataset, Model, and Benchmark for Health Management at Home. ICML 2026

C26S Xiao, J Fu, Z Xie*, L Shou. TokenTiming: A Dynamic Alignment Method for Universal Speculative Decoding Model Pairs. ACL 2026 Main Oral & Awards Nomination

C25F Lin, C You, Z Xie, Z Luo, M Zhang. SaCal: An Efficient Saliency-Guided Causal Framework for Interpretable Healthcare Analytics. ICDE 2026 (To Appear)

C24J Song, Y Liu, G Hu, Z Xie, M Yang, BC Ooi, K Zhou. FAVOR: Efficient Filter-Agnostic Vector ANNS Based on Selectivity-Aware Exclusion Distances. SIGMOD 2026

C23T Lin, et al.. OmniCT: Towards a Unified Slice-Volume LVLM for Comprehensive CT Analysis. ICLR 2026

C22C Lv, H Li, D Yang, Z Xie, L Chen, CS Jensen. DeXOR: Enabling xor in Decimal Space for Streaming Lossless Compression of Floating-point Data. VLDB 2026

C21Y Peng, D Yang, Z Xie, J Sun, L Shou, K Chen, G Chen. SVFusion: A CPU-GPU Co-Processing Architecture for Large-Scale Real-Time Vector Search. VLDB 2026 Industry Collaboration

C20Y Wu, et al.. SafeLoad: Efficient Admission Control Framework for Identifying Memory-Overloading Queries in Cloud Data Warehouses. VLDB 2026 Industry Collaboration

C19S Wu, et al.. MorphingDB: A Task-Centric AI-Native DBMS for Model Management and Inference. SIGMOD 2026

Journal

J5G Chen, et al.. Generative AI for Healthcare: Fundamentals, Challenges, and Perspectives. SCIENCE CHINA Information Sciences

J4X Chen, Z Xie*, H Li, K Chen, L Shou, D Jiang, G Chen. PIMSHARE: Scheduling for Multi-DNN Inference on Processing-in-memory Accelerated Edge Server. IEEE TCAD 2026

Conference

C18H Lin, S Wan, Z Xie*, K Chen*, M Zhang, L Shou, G Chen. A Comprehensive Study of Shapley Value in Data Analytics. VLDB 2025

C17G Hu, S Cai, TTA Dinh, Z Xie*, C Yue, G Chen, BC Ooi. HAKES: Scalable Vector Database for Embedding Search Service. VLDB 2025

C16Z Ji, X Wang, Z Luo, Z Xie, M Zhang. Optimized Batch Prompting for Cost-effective LLMs. VLDB 2025

C15Y Zhou, Z Li, J Zhang, J Wang, Y Wang, Z Xie*, K Chen, L Shou*. FloE: On-the-Fly MoE Inference on Memory-constrained GPU. ICML 2025

C14Y Peng, Z Xie*, K Chen*, G Chen, L Shou. Towards Automatic and Efficient Prediction Query Processing in Analytical Database. ICDE 2025

Journal

J3P Lu, Z Xie*, D Jiang, K Chen, L Shou. Cohort query processing without misleading aging effects. VLDB Journal

J2J Zhang, J Wang, H Li, Z Xie, K Chen, L Shou. CHASe: Client Heterogeneity-Aware Data Selection for Effective Federated Active Learning. TKDE 2025

Conference

C13Z Ji, Z Xie, Y Wu, M Zhang. LBSC: A Cost-Aware Caching Framework for Cloud Databases. ICDE 2024 Best Runner-Up Paper

Conference

C12C Yue, TTA Dinh, Z Xie, M Zhang, G Chen, BC Ooi, X Xiao. GlassDB: An efficient verifiable ledger database system through transparency. VLDB 2023

C11Y Ma, Z Xie, J Wang, K Chen, L Shou. Continual Federated Learning Based on Knowledge Distillation. IJCAI 2022

C10J Zhang, S Wu, J Zhao, Z Xie, F Li, Y Gao, G Chen. A sampling-based learning framework for big databases. WWW 2022 Industry Collaboration

C9M Zhang, Z Xie, C Yue, Z Zhong. Spitz: A verifiable database system. VLDB 2020

C8C Yue, Z Xie, M Zhang, G Chen, BC Ooi, S Wang, X Xiao. Analysis of indexing structures for immutable data. ACM SIGMOD 2020

C7Z Xie, H Ying, C Yue, M Zhang, G Chen, BC Ooi. Cool, a COhort OnLine analytical processing system. ICDE 2020

C6Z Xie, Q Cai, F He, GY Ooi, W Huang, BC Ooi. Cohort analysis with ease. ACM SIGMOD 2018 Demo.

C5Z Xie, Q Cai, G Chen, R Mao, M Zhang. A comprehensive performance evaluation of modern in-memory indices. ICDE 2018

C4Q Cai, Z Xie, M Zhang, G Chen, HV Jagadish, BC Ooi. Effective temporal dependence discovery in time series data. VLDB 2018

C3S Wang, TTA Dinh, Q Lin, Z Xie, M Zhang, Q Cai, G Chen, W Fu, BC Ooi, P Ruan. Forkbase: An efficient storage engine for blockchain and forkable applications. VLDB 2018

C2Z Xie, Q Cai, HV Jagadish, BC Ooi, WF Wong. Parallelizing skip lists for in-memory multi-core database systems. ICDE 2017

C1BC Ooi, et al.. SINGA: A distributed deep learning platform. ACM MM 2015

Journal

J1Q Cai, C Cui, Y Xiong, W Wang, Z Xie, M Zhang. A survey on deep reinforcement learning for data processing and analytics. TKDE 2022

Awards and Honors

• April 2026 ACL Awards Nomination
• Sep 2025 CCF Science and Technology Achievement Award - Natural Science Award (Second Prize)
• June 2024 SIGMOD Systems Award
• April 2024 ICDE Best Runner-Up Paper
• Aug 2024 VLDB Distinguished Reviewer

Educations

I fondly remember and honor my beloved supervisor, Professor Ooi Beng Chin. He is a brilliant scholar, a leading figure in dataXAI research, and a respected mentor and friend. His academic spirit, dedication, and kindness will remain with us always.

• 2014.08 - 2020.01, National University of Singapore, Ph.D., supervised by Prof. Beng Chin OOI.
• 2010.09 - 2014.06, Shanghai Jiao Tong University, Bachelor, supervised by Prof. Bin YAO.

Professional Activities

• VLDB Review Board/PC Member: 2022, 2023, 2024, 2026
• ICDE PC Member: 2023, 2027
• EDBT PC Member: 2026
• CIKM PC Member: 2025, 2026

Students

I am looking for highly self-motivated Ph.D. and master’s students, as well as undergraduate research interns. If you are interested, please feel free to email me your CV.