Xiangyu Zhao is a tenure-track assistant professor of Data Science at City University of Hong Kong (CityU). Prior to CityU, he completed his Ph.D. under the advisory of Prof. Jiliang Tang at DSE Lab of MSU, his M.S. under the advisory of Prof. Enhong Chen at BDAA Lab of USTC, and his B.Eng. under the advisory of Prof. Ming Tang and Prof. Tao Zhou at BigData Center of UESTC.
His current research interests include data mining and machine learning, especially
Personalization, Recommender System, Search Engine, Online Advertising, and Information Retrieval
Large Language Model, AGI, AutoML, Reinforcement Learning, Graph Learning, Trustworthy AI, and Multimodal ML
AI + X: Urban Computing & Smart City, Healthcare, Education, Carbon Neutral, Social Computing, Finance, and Ecosystem
His research has been awarded ICDM'22 and ICDM'21 Best-ranked Papers, Global Top 100 Chinese New Stars in AI (Top 25 in Data Mining), CCF-Tencent Open Fund (twice), CCF-Ant Research Fund, CCF-BaiChuan-Ebtech Foundation Model Fund, Ant Group Research Fund, Tencent Focused Research Fund, Criteo Faculty Research Award, Bytedance Research Collaboration Program, MSU Dissertation Fellowship, and nomination for Joint AAAI/ACM SIGAI Doctoral Dissertation Award. He is the awardee of the HK RGC Research Impact Fund with a grant of HK$7 million. He serves as top data science conference (senior) program committee members, session chairs and journal reviewers. He serves as the organizers of DRL4KDD and DRL4IR workshops at KDD'19, WWW'21, SIGIR'20/21/22 and CIKM'23, and a lead tutor at KDD'23, WWW'21/22/23, IJCAI'21/23 and WSDM'23. He also serves as the founding academic committee member of MLNLP, the largest Chinese AI community with millions of subscribers. The models and algorithms from his research have been launched in the online systems of various companies, such as Amazon, Google, Facebook, Linkedin, Criteo, Lyft, Baidu, Tencent, Ant Group, Kuaishou, JD.com, and Bytedance.
Openings
- I have PhD (ddl: Jan 15, 2024) and Joint-PhD (双一流A类+国科大, ddl: Dec 8, 2023) positions every year
- Postdoc, Self-financed PhD, Part-time PhD, RA and Visiting Students positions are open year round
- Please click HERE for more details (CityU is ranked around #50 in QS World University Rankings 2016-2023)
Our Large Language Models (LLMs) Papers
- When MOE Meets LLMs: Parameter Efficient Fine-tuning for Multi-task Medical Applications (SIGIR'2024) [link]
- MILL: Mutual Verification with Large Language Models for Zero-Shot Query Expansion (NAACL'2024) [link]
- Large Multimodal Model Compression via Efficient Pruning and Distillation (WWW'2024, Oral Presentation) [link]
- Multi-perspective Improvement of Knowledge Graph Completion with Large Language Models (COLING'2024) [link]
- Recommender Systems in the Era of Large Language Models (LLMs) (IEEE TKDE) [link]
- A Unified Framework for Multi-Domain CTR Prediction via Large Language Models [link]
- Rethinking Large Language Model Architectures for Sequential Recommendations [link]
- Large Language Model Distilling Medication Recommendation Model [link]
- Agent4Ranking: Semantic Robust Ranking via Personalized Query Rewriting Using Multi-agent LLM [link]
- Tired of Plugins? Large Language Models Can Be End-To-End Recommenders [link]
- E4SRec: An Elegant Effective Efficient Extensible Solution of Large Language Models for Sequential Recommendation [link]
- Editing Factual Knowledge and Explanatory Ability of Medical Large Language Models [link]
- Large Language Models for Generative Information Extraction: A Survey [link]
Our Research Highlights
- WWW'2024, Oral Presentation (10%)
Large Multimodal Model Compression via Efficient Pruning and Distillation [link]
- ICDM'2022, Best-ranked Papers Award
AutoAssign: Automatic Shared Embedding Assignment in Streaming Recommendation [link]
- IJCAI'2022, Long Oral Presentation (3.75%)
MLP4Rec: A Pure MLP Architecture for Sequential Recommendations [link]
- WSDM'2021, Top-4 Most Cited Paper 4 / 155 Accepted Papers
Towards Long-term Fairness in Recommendation [link]
- ICDM'2021, Best-ranked Papers Award, Top-10 Most Cited Paper 10 / 208 Accepted Papers
AutoEmb: Automated Embedding Dimensionality Search in Streaming Recommendations [link]
- ACM SIGWEB Newsletter, Top-2 Most Cited Paper 2 / 551 in History
Deep Reinforcement Learning for Search, Recommendation, and Online Advertising: A Survey [link]
- RecSys'2018, Top-1 Most Cited Paper 1 / 117 Accepted Papers
Deep Reinforcement Learning for Page-wise Recommendations [link]
- KDD'2018, Top-20 Most Cited Paper 19 / 294 Accepted Papers
Recommendations with Negative Feedback via Pairwise Deep Reinforcement Learning [link]
- CIKM'2017, Top-20 Most Cited Paper 20 / 350 Accepted Papers
Modeling Temporal-Spatial Correlations for Crime Prediction [link]
Our Survey Papers
- Embedding in Recommender Systems: A Survey [link]
- Recommender Systems in the Era of Large Language Models (LLMs) (IEEE TKDE) [link]
- Large Language Models for Generative Information Extraction: A Survey [link]
- Automated Machine Learning for Deep Recommender Systems: A Survey (ACM TORS) [link]
- Multi-Task Deep Recommender Systems: A Survey [link]
- Multimodal Recommender Systems: A Survey [link]
- A Comprehensive Survey on Trustworthy Recommender Systems [link]
- Deep Reinforcement Learning for Search, Recommendation, and Online Advertising: A Survey (ACM SIGWEB Newsletter) [link]
- Mobile Trajectory Anomaly Detection: Taxonomy, Methodology, Challenges, and Directions (IEEE IoT) [link]
- Crime in Urban Areas: A Data Mining Perspective (ACM SIGKDD Explorations) [link]
- On the Opportunities of Green Computing: A Survey [link]
Call for Paper
- NLPCC'24 [Website]
- Tutorial Proposal @ APWeb-WAIM'24 [Website]
- AgentIR: 1st Workshop on Agent-based Information Retrieval @ SIGIR'24 [Website]
- 2nd Recommendation With Generative Models @ WWW'24 [Website]