Hello! This is Zhifu Wei (魏智富 in Chinese). I am an undergraduate student at the School of Computer Science and Engineering of Northeastern University, Shenyang, China. I am fortunate to be advised by Prof. Guibing Guo, with whom I work on sequential recommendation and long-tail recommendation.
My research interests focus on large language models (LLMs), AI agents, and recommender systems. I am particularly interested in how LLMs can strengthen the representation of sparse (tail) items in sequential recommendation, and how multimodal agents can safely and reliably operate in real-world workflows.
Please feel free to contact me by email if you have any questions or are seeking collaborations.
你好！我是魏智富，目前是东北大学计算机科学与工程学院本科生，现居沈阳。我很荣幸师从郭贵冰教授，从事序列推荐与长尾推荐方向的研究。
我的研究兴趣集中在大语言模型（LLM）、AI Agent 与推荐系统。我特别关注如何利用大语言模型增强序列推荐中长尾物品的表征质量，以及多模态智能体如何在真实工作流中安全可靠地运作。
如果你对我的研究感兴趣，或希望开展交流与合作，欢迎通过邮件与我联系。
🔥 News
- 2026.01: 🎉 Our paper on LLM-enhanced tail-item sequential recommendation (FAERec) was accepted at SIGIR 2026.
- 2026.01: 🏅 I am honored to receive the full travel scholarship for the AAAI-26 Undergraduate Consortium.
- 2025.11: 🎉 Our paper on tail-aware data augmentation (TADA) was accepted at WWW 2026.
🔥 新闻
- 2026.01: 🎉 我们基于大语言模型的长尾序列推荐论文（FAERec）被 SIGIR 2026 接收。
- 2026.01: 🏅 我很荣幸获得 AAAI-26 本科生联盟全额差旅奖学金。
- 2025.11: 🎉 我们基于尾部感知数据增强的论文（TADA）被 WWW 2026 接收。
📝 Publications
† indicates corresponding author.
-
Fusion and Alignment Enhancement with Large Language Models for Tail-item Sequential Recommendation
Zhifu Wei, Yizhou Dang, Guibing Guo†, Chuang Zhao, Zhu Sun†
International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2026) · Code · arXiv -
TADA: Tail-Aware Data Augmentation for Long-Tail Sequential Recommendation
Yizhou Dang, Zhifu Wei, Minhan Huang, Lianbo Ma, Jianzhe Zhao, Guibing Guo†, Xingwei Wang†
The ACM Web Conference (WWW 2026) · Code · arXiv -
A Realistic Benchmark for Multi-Interface Safety-Aware Computer-Use Agents in Clinical Workflows
Yushi Feng, Zhifu Wei, Ziyi He, Pui Hang Leung, Peixiang Huang, Yunhai Wang, Weifa Yang, Lequan Yu
Manuscript in preparation
📝 论文
† 表示通讯作者。
-
基于大语言模型融合与对齐增强的长尾序列推荐
魏智富，党翌洲，郭贵冰†，赵闯，朱顺†
ACM SIGIR 国际信息检索大会（SIGIR 2026） · 代码 · arXiv -
TADA：面向长尾序列推荐的尾部感知数据增强
党翌洲，魏智富，黄敏涵，马连博，赵健哲，郭贵冰†，王兴伟†
The ACM Web Conference（WWW 2026） · 代码 · arXiv -
医疗工作流中面向多接口的安全感知计算机操作智能体基准
冯煜石，魏智富，何子怡，Pui Hang Leung，黄培轩，王云海，杨威法，于乐全
Manuscript in preparation（撰写中）
📖 Education
- Sep. 2023 – Jun. 2027 (Expected), B.E. in Computer Science and Technology, Northeastern University, Shenyang, China.
📖 教育经历
- 2023.09 – 2027.06（预计），东北大学计算机科学与技术专业工学学士，沈阳，中国。
🏆 Honors and Awards
- 2026.01: 🏅 Full Travel Scholarship, AAAI-26 Undergraduate Consortium (AAAI-UC)
🏆 荣誉奖励
- 2026.01: 🏅 AAAI-26 本科生联盟（AAAI-UC）全额差旅奖学金
💼 Research Experience
- Research on Long-Tail Recommendation — Aug. 2025 – Jan. 2026
- Research Assistant, advised by Prof. Guibing Guo
- Investigated representation learning and data augmentation for long-tail sequential recommendation, co-authoring two papers accepted at top conferences: FAERec (SIGIR 2026), which combines collaborative signals with LLM-derived semantics to better represent sparse tail items, and TADA (WWW 2026), a tail-aware data augmentation framework for sparse user-item interactions.
- Computer-Use Agents in Clinical Workflows (OS-MedWorld) — Feb. 2026 – May 2026
- Research Assistant, advised by Prof. Lequan Yu, The University of Hong Kong
- Built OS-MedWorld, a realistic and resettable clinical benchmark for computer-use agents, and evaluated frontier VLM agents on risk recognition and multi-step task execution.
💼 科研经历
- 长尾推荐研究（FAERec & TADA） — 2025.08 – 2026.01
- 研究助理，导师：郭贵冰教授
- 研究长尾序列推荐的表示学习与数据增强，作为主要贡献者参与两项被顶级会议接收的工作：FAERec（SIGIR 2026），融合协同信号与 LLM 语义以增强稀疏长尾物品的表征；TADA（WWW 2026），面向稀疏交互的尾部感知数据增强框架。
- 医疗工作流中的计算机操作智能体（OS-MedWorld） — 2026.02 – 2026.05
- 研究助理，导师：于乐全教授，香港大学
- 搭建 OS-MedWorld 可重置临床基准环境，评测前沿视觉语言智能体在风险识别与多步任务执行中的表现。
🎤 Conference Participation
- 2026.01: Presented at the AAAI-26 Undergraduate Consortium (full travel scholarship).
🎤 参会经历
- 2026.01: 参加 AAAI-26 本科生联盟并作报告（全额差旅资助）。
🛠 Skills
- Programming: Python, C/C++, PyTorch, NumPy
- Research & Engineering Tools: Linux, Docker, Git, Overleaf
🛠 技能
- 编程语言： Python、C/C++、PyTorch、NumPy
- 研究与工程工具： Linux、Docker、Git、Overleaf