Short Bio

Dr. Xin Zhang obtained his PhD degree from The Hong Kong University of Science and Technology (HKUST) in 2026, under the supervision of Professor Lei Chen (ACM Fellow & IEEE Fellow). He received his Bachelor’s degree from Peking University (PKU) in 2020, where he was advised by Professor Bin Cui (IEEE Fellow).

Research Interest

Dr. Xin Zhang’s research focuses on building efficient Machine Learning Systems tailored for modern deep learning workloads, including Large Language Models (LLMs), Graph Neural Networks (GNNs), and Recommendation Systems. His work spans the end-to-end model lifecycle, addressing core challenges in both scalable training and efficient serving.

His current work focuses on LLM inference optimization, improving KV cache management in online serving, chunk reuse for RAG service, and batching/scheduling for semantic operators. His prior research centered on scaling GNN and Recommendation model training through GPU caching, heterogeneous workload dispatching, and optimized data sampling.

Publications

SLO-Driven Dual-Budget Scheduling for Cache-Enabled RAG Serving.
Hongbo Yin, Xin Zhang, Jingzhi Fang, Yanyan Shen, Lei Chen.
Accepted by ICDE 2027.

Efficient GNN Training on Giant Graphs with Collective Batching and Scheduling.
Xin Zhang, Yanyan Shen, Yingxia Shao, Haoyang Li, Lei Chen.
PVLDB 2026. [paper] [code]

Graph Neural Network Training: From Data Management Perspective
Springer Nature 2026. [book]

RelServe: Fast LLM Inference Serving on Relational Data.
Xin Zhang, Shihong Gao, Yanyan Shen, Haoyang Li, Lei Chen.
Preprint 2025. [paper]

Apt-Serve: Adaptive Request Scheduling on Hybrid Cache for Scalable LLM Inference Serving.
Shihong Gao, Xin Zhang, Yanyan Shen, Lei Chen.
SIGMOD 2025. [paper] [code]

Efficient Training of Graph Neural Networks on Large Graphs.
Yanyan Shen, Lei Chen, Jingzhi Fang, Xin Zhang, Shihong Gao, Hongbo Yin.
PVLDB 2024. [paper] [code]

SIMPLE: Efficient Temporal Graph Neural Network Training at Scale with Dynamic Data Placement.
Shihong Gao, Yiming Li, Xin Zhang, Yanyan Shen, Yingxia Shao, Lei Chen.
SIGMOD 2024. [paper] [code]

DUCATI: A Dual-Cache Training System for Graph Neural Networks on Giant Graphs with the GPU.
Xin Zhang, Yanyan Shen, Yingxia Shao, Lei Chen.
SIGMOD 2023. [paper] [code]

Feature-Oriented Sampling for Fast and Scalable GNN Training.
Xin Zhang, Yanyan Shen, Lei Chen.
ICDM 2022. [paper] [code]

HET-GMP: A Graph-based System Approach to Scaling Large Embedding Model Training.
Xupeng Miao, Yining Shi, Hailin Zhang, Xin Zhang, Xiaonan Nie, Zhi Yang, Bin Cui.
SIGMOD 2022. [paper] [code]

Machine Reading Comprehension: a Literature Review.
Xin Zhang, An Yang, Sujian Li, Yizhong Wang.
Preprint 2019. [paper]

Services

Invited journal reviewer:

• TPAMI (IEEE Transactions on Pattern Analysis and Machine Intelligence) 2026
• TPDS (IEEE Transactions on Parallel and Distributed Systems) 2026
• TC (IEEE Transactions on Computers) 2026
• TSC (IEEE Transactions on Services Computing) 2026
• TMC (IEEE Transactions on Mobile Computing) 2026
• TMM (IEEE Transactions on Multimedia) 2026
• TAC (IEEE Transactions on Affective Computing) 2026
• FCS (Frontiers of Computer Science) 2026

Last Update: September 2026