Ziwei Liu

Ziwei Liu

City University of Hong Kong

Generate CV PDF
WeChat
WeChat QR code for Ryan_2701
ID: Ryan_2701

Global Visiting Record

0 总访问量 0 已定位国家 0 本次计入
1-10 访问
11-50 访问
50+ 访问

最近访问者

Currently, I am a first-year Ph.D. student in the AML Lab at City University of Hong Kong, supervised by Prof. Zhao Xiangyu. Before this, I earned my Bachelor’s degree in Robotics Engineering from Southeast University, where Prof. Gan Yahui and Prof. Li Jun co-supervised me.

My research interests include Recommender Systems, Information Retrieval, and Large Language Models. I have published several papers at top international AI conferences, with 82 total Google Scholar citations.

🔥 News

Latest updates · scroll for earlier news ↓

  • One paper was accepted by RecSys 2026. Congratulations to Yuxuan!
  • I received the KDD 2026 Student Travel Award.
  • Our tutorial, “Tutorial on Generative Recommendation: Foundations and Frontiers,” was accepted by KDD 2026. Many thanks to Xiaopeng for leading this work!
  • Our paper H²Rec was accepted by KDD 2026 ADS Track (CCF-A). It has been deployed on RedNote (Xiaohongshu), achieving ADVV +0.89% and COST +0.59%. Many thanks to Yejing and Qidong!
  • Our new paper ComeIR was released on arXiv. It identifies a representation bottleneck in decoupled two-stage generative recommendation and introduces Engram memory for improved input construction and decoding.
  • Our paper LLM-EDT was accepted by SIGIR 2026 (CCF-A). Many thanks to Qidong and Yejing!
  • One paper was accepted by WWW 2026 (CCF-A). Many thanks to Yejing!
  • Our new paper H²Rec was released on arXiv. It harmonizes Semantic IDs with Hash IDs to address collaborative overwhelming.
  • Our new paper LLM-EDT was released on arXiv. It introduces a general cross-domain sequential recommender for domain imbalance and rough profiling.
  • I received the AAAI 2025 Travel Award.
  • Our paper SIGMA: Selective Gated Mamba for Sequential Recommendation was accepted by AAAI 2025 (CCF-A). Many thanks to Qidong for leading this work!

📝 First/Co-first Author Publications

Overview of the ComeIR framework
ComeIR reconstructs item-aware inputs and restores token-level evidence with dual-level conditional memory.

Conditional Memory Enhanced Item Representation for Generative Recommendation

Ziwei Liu*, Yejing Wang*, Shengyu Zhou, Xinhang Li, Xiangyu Zhao

Preprint, 2026

Paper Code coming soon

TL;DR: ComeIR uses token scoring and dual-level Engram memories to preserve item identity and SID structure, then reuses the memories during decoding to bridge item-level inputs and token-level generation.

Overview of the H2Rec framework
H²Rec harmonizes multi-granular Semantic IDs with unique collaborative Hash IDs through dual-branch modeling.

The Best of Both Worlds: Harmonizing Semantic and Hash IDs for Sequential Recommendation

Ziwei Liu*, Yejing Wang*, Wanyu Wang, Zejian Wang, Qidong Liu, Zijian Zhang, Wei Huang, Chong Chen, Xiangyu Zhao

ACM SIGKDD Conference on Knowledge Discovery and Data Mining, ADS Track (KDD), 2026 · CCF-A

Paper Code

TL;DR: H²Rec combines the semantic generalization of SIDs with the collaborative uniqueness of HIDs, balancing recommendation quality for both head and long-tail items.

Overview of the LLM-EDT framework
LLM-EDT couples transferable item augmentation, dual-phase training, and domain-aware user profiling.

LLM-EDT: Large Language Models Enhanced Cross-domain Sequential Recommendation with Dual-phase Training

Ziwei Liu*, Qidong Liu*, Wanyu Wang, Yejing Wang, Pengyue Jia, Tong Xu, Wei Huang, Chong Chen, Xiangyu Zhao

ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR), 2026 · CCF-A

Paper Code

TL;DR: LLM-EDT addresses domain imbalance, noisy augmentation, and rough profiling via transferable item augmentation, domain-specific fine-tuning, and adaptive domain-aware preference aggregation.

Overview of the SIGMA framework
SIGMA enhances Mamba with partial bidirectional modeling, selective gating, and short-term feature extraction.

SIGMA: Selective Gated Mamba for Sequential Recommendation

Ziwei Liu*, Qidong Liu*, Yejing Wang, Wanyu Wang, Pengyue Jia, Maolin Wang, Zitao Liu, Yi Chang, Xiangyu Zhao

AAAI Conference on Artificial Intelligence (AAAI), 2025 · CCF-A

Paper Code

TL;DR: SIGMA equips Mamba with partially flipped bidirectional modeling, an input-dependent selective gate, and a feature-extracting GRU for efficient long- and short-term sequential recommendation.

💬 Tutorials

🎖 Honors and Awards

  • 2026.06 · KDD 2026 Student Travel Award and Volunteer
  • 2025.01 · AAAI 2025 Travel Award and Volunteer
  • 2024.12 · Silver Medal, Kaggle LLM Prompt Recovery (44/2,175)
  • 2023.06 · Excellent Graduation Project, Southeast University

🏷️ Service

Conferences

  • ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2027), ADS Track
  • AAAI Conference on Artificial Intelligence (AAAI 2027), Main Track
  • ACM International Conference on Multimedia (MM 2026), Main and DB Tracks
  • ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2026)
  • AAAI Conference on Artificial Intelligence (AAAI 2026), Main and AIA Tracks
  • ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2025)
  • ACM Conference on Recommender Systems (RecSys 2025)

Journals

  • IEEE Transactions on Knowledge and Data Engineering (TKDE)
  • ACM Transactions on Knowledge Discovery from Data (TKDD)
  • ACM Transactions on Information Systems (TOIS)

📖 Education

  • 2025.09–Present · Ph.D. in Data Science, City University of Hong Kong
  • 2023.09–2024.10 · M.E. in Data Science, City University of Hong Kong
  • 2019.09–2023.06 · B.E. in Robotics Engineering, Southeast University

💻 Experience

  • 2024.05–2024.12 · Research Assistant, Chinese University of Hong Kong, Shenzhen