Sheng Zhang is a Ph.D. student at City University of Hong Kong (CityUHK) under the supervision of Prof. Xiangyu Zhao. Prior to his master's study, he received his M.Sc. degree at CityUHK under the guidance of Prof. Xiangyu Zhao, and received his B.Sc. degree from the University of Electronic Science and Technology of China (UESTC) under the guidance of Prof. Jie Huang. When working at the High Magnetic Field Laboratory of the Chinese Academy of Sciences, his research focused on the application of large language models and data-driven optimization of magnet array design. Currently, his research focus on Large Language Model (LLM) Agents, Sequential Recommender Systems, and AI for Science.
LLM Agents
Sequential Recommender Systems
AI for Science
๐ฅ News
- 2026.08: ย ๐๐ Our paper, R$^2$-Searcher: Calibrating Retrieval and Reasoning Boundaries for Agentic Search, has been accepted by CIKM 2026!
- 2026.07: ย ๐๐ Our paper, Curriculum Learning for Efficient Chain-of-Thought Distillation via Structure-Aware Masking and GRPO is accepted by SDM 2026! Congratulations to Bowen!
- 2026.05: ย ๐๐ Our Tutorial, Tutorial on Generative Recommendation: Foundations and Frontiers is accepted by KDDโ26! Congratulations to Xiaopeng!
- 2026.05: ย ๐๐ Our paper, Towards Pareto-Optimal Tool-Integrated Agents with Pareto Ranking Policy Optimization, has been accepted by ICML 2026 Spotlight! Congratulations to Junyi!
- 2026.04: ย ๐๐ Our paper, MemSearch-o1: Empowering Large Language Models with Reasoning-Aligned Memory Growth in Agentic Search, has been accepted by ACL 2026!
- 2026.03: ย ๐๐ Our paper, Embedding in recommender systems: A survey, has been accepted by TOIS.
- 2025.11: ย ๐๐ Our paper, Renormalization Group Guided Tensor Network Structure Search, has been accepted by AAAI 2026.
- 2025.10: ย ๐๐ Our paper, Embedding in Recommender Systems: A Survey has been updated in Google Scholar.
- 2025.05: ย ๐๐ Our paper FindRec: Stein-Guided Entropic Flow for Multi-Modal Sequential Recommendation is accepted by KDD 2025.
- 2025.04: ย ๐๐ Our paper DANCE: Resource-Efficient Neural Architecture Search with Data-Aware and Continuous Adaptation is accepted by IJCAI 2025.
- 2025.03: ย ๐๐ Our paper STAR-Rec: Making Peace with Length Variance and Pattern Diversity in Sequential Recommendation is accepted by SIGIR 2025.
- 2024.11: ย ๐๐ Our paper GLINT-RU: Gated Lightweight Intelligent Recurrent Units for Sequential Recommender Systems is accepted by KDD 2025.
- 2024.11: ย ๐๐ I am awarded the Outstanding Academic Performance Award (CGPA Top 1 Student) and Outstanding Research Project Award by the Department of Data Science at City University of Hong Kong.
- 2024.07: ย ๐๐ Our paper DNS-Rec: Data-aware Neural Architecture Search for Recommender Systems is accepted by Recsys 2024.
- 2024.06: ย ๐๐ I am awarded Master of Science in Data Science with Distinction.
๐ Publications
ACL 2026 ยท CCF-A
MemSearch-o1: Empowering Large Language Models with Reasoning-Aligned Memory Growth in Agentic Search
MemSearch-o1 dynamically grows fine-grained memory fragments from memory seed tokens from the queries, and reorganizes a globally connected memory path for agentic search. This shifts the paradigm of memory management to structured, token-level growth with path-based reasoning.
KDD 2025 ยท CCF-A
GLINT-RU: Gated Lightweight Intelligent Recurrent Units for Sequential Recommender Systems
GLINT-RU is a lightweight and efficient SRS leveraging a single-layer dense selective Gated Recurrent Units (GRU) module to accelerate inference and generate high-quality latent representations. GLINT-RU achieves at least 15% inference speed improvement over efficient recommender system such as Mamba4Rec, LinRec while improving the accuracy simultaneously.
CIKM 2026 ยท CCF-B
R2-Searcher: Calibrating Retrieval and Reasoning Boundaries for Agentic Search
We propose Rยฒ-Searcher, a novel framework that explicitly explores and calibrates the retrieval and reasoning bound aries via fine-grained, query-token-guided evidence modeling and post-retrieval reflection. The tree-based reinforcement learning strategy is implemented to jointly optimize retrieval and reasoning boundaries to enhance the process-level rewards.
SIGIR 2025 ยท CCF-A
STAR-Rec: Making Peace with Length Variance and Pattern Diversity in Sequential Recommendation
We theoretically demonstrate how the state space model and attention mechanisms can be naturally unified in recommendation scenarios, where SSM captures temporal dynamics through state compression while attention models both similar and diverse item relationships. STAR-Rec consistently improves both accuracy and efficiency, particularly in scenarios involving diverse user behaviors and varying sequence lengths.
RecSys 2024 ยท CCF-B
DNS-Rec: Data-aware Neural Architecture Search for Recommender Systems
DNS-Rec is specifically designed to tailor compact network architectures for attention-based SRS models, thereby ensuring accuracy retention. It incorporates data aware gates to enhance the performance of the recommendation network, and employs a dynamic resource constraint strategy for neural achitecture search.
Other Publications
- Junyi Li, Xiaowei Qian, Yingyi Zhang, Wenlin Zhang, Guojing Li, Sheng Zhang, Xiao Han, Yichao Wang, Xiangyu Zhao. Towards Pareto-Optimal Tool-Integrated Agents with Pareto Ranking Policy Optimization. Forty-Third International Conference on Machine Learning (ICML Spotlight) (CCF-A)
- Bowen Yu, Sheng Zhang, Binhao Wang, Yi Wen, Jingtong Gao, Bowen Liu, Zimo Zhao, Wanyu Wang, Maolin Wang, Xiangyu Zhao. (2026) Curriculum Learning for Efficient Chain-of-Thought Distillation via Structure-Aware Masking and GRPO. In SIAM International Conference on Data Mining (SDM) (CCF-B)
- Maolin Wang, Bowen Yu, Sheng Zhang, Linjie Mi, Wanyu Wang, Yiqi Wang, Pengyue Jia, Xuetao Wei, Zenglin Xu, Ruocheng Guo, Xiangyu Zhao. (2026). Renormalization Group Guided Tensor Network Structure Search. In Proceedings of the 40th AAAI Conference on Artificial Intelligence (AAAI) (CCF-A).
- Maolin Wang, Tianshuo Wei, Sheng Zhang, Ruocheng Guo, Wangyu Wang, Shanshan Ye, Lixin Zou, Xuetao Wei, Xiangyu Zhao. (2025). DANCE: Resource-Efficient Neural Architecture Search with Data-Aware and Continuous Adaptation. In International Joint Conference on Artificial Intelligence (IJCAI) (CCF-B).
- Maolin Wang, Xinjian Zhao, Wanyu Wang, Sheng Zhang, Jiansheng Li, Bowen Yu, Binhao Wang, Shucheng Zhou, Dawei Yin, Qing Li, Ruocheng Guo, Xiangyu Zhao. Embedding in Recommender Systems: A Survey. ACM Transactions on Information Systems (TOIS) (CCF-A)
- Maolin Wang, Yutian Xiao, Binhao Wang, Sheng Zhang, Shanshan Ye, Wanyu Wang, Hongzhi Yin, Ruocheng Guo, Zenglin Xu. (2025). FindRec: Stein-Guided Entropic Flow for Multi-Modal Sequential Recommendation. In SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) (CCF-A).
๐ฌ Tutorial
- Tutorial on Generative Recommendation: Foundations and Frontiers (GitHub) (KDD 2026, CCF-A).
โ Services
- Student Committee of MLNLP Community.
- Program Committee of the 20th ACM International Conference on Web Search and Data Mining (WSDMโ2027).
- Program Committee of AAAI Association for the Advanced Artificial Intelligence (AAAIโ2027 main track).
- ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDDโ2027 AI for science track).
- ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIRโ2026 main track and reproducibility track).
- ACM International Conference on Multimedia (MMโ2026 main track)
- ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDDโ2026 AI for science track and datasets and benchmarks track).
- International World Wide Web Conference (WWWโ2026)
- AAAI Association for the Advanced Artificial Intelligence (AAAIโ2026 main track).
- IEEE Transactions on Knowledge and Data Engineering (TKDE).
๐ Honors and Awards
- Master of Science in Data Science Outstanding Research Project Award (2024.11)
- Master of Science in Data Science Outstanding Academic Performance Award (2024.11)
- Master of Science in Data Science with Distinction (2024.06)
- Model Student Scholarship of UESTC for two consecutive years (2020 - 2021)
๐ Educations
2025.09 - 2029.06
Ph.D. in Data Science
Department of Data Science, City University of Hong Kong
Hong Kong SAR, China ยท Supervisor: Prof. Xiangyu Zhao
2023.09 - 2024.06
M.Sc. in Data Science
Department of Data Science, City University of Hong Kong
Hong Kong SAR, China ยท Supervisor: Prof. Xiangyu Zhao
2019.09 - 2023.06
B.Sc. in Data Science and Big Data Technology
Department of Mathematical Sciences, University of Electronic Science and Technology of China
Chengdu, China ยท Supervisor: Prof. Jie Huang
๐ป Work Experience
Assistant Engineer