PhD Student in Computer Science · 2027-
Wei Li (William)
Shanghai Jiao Tong University × Beijing Zhongguancun Academy
My undergraduate education is in Computer Science and Technology at Shanghai University (2023-2027). I am a PhD student in Computer Science at Shanghai Jiao Tong University × Beijing Zhongguancun Academy (2027-), advised by Prof. Guanjie Zheng and Prof. Jie Feng. My research focuses on machine learning for structured and temporal data, including representation learning, forecasting, generative modeling, clustering, and AI research systems.
News
- Awarded the National Scholarship and the Baosteel Scholarship.
- SDFlow accepted for Poster Presentation at NeurIPS 2026 (CCF-A).
- The APCL (KDD 2026, CCF-A) blog post is now live.
- APCL scheduled for Poster Presentation at KDD 2026 (CCF-A).
- One full paper (APCL) accepted to KDD 2026 (CCF-A).
- Won International First Prize at ASC26.
- One full paper (ClusterPatchTST) accepted for Oral Presentation at DASFAA 2026 (CCF-B).
- Advanced to the finals of the ASC Student Supercomputer Challenge.
- Two papers (ScatterFusion, AWGFormer) accepted for Poster Presentation at ICASSP 2026 (CCF-B).
- Awarded the Xiaomi Scholarship.
- Awarded the Shanghai Scholarship after a year that included ranking #1 in GPA, publishing four first-author papers at CCF-C ranked conferences, and winning multiple national programming contest awards.
- Awarded First Prize in the Baidu Star Programming Contest - Talent Track and MaTiBei Collegiate Programming Contest.
- Two papers (SWIFT, TimeFlowDiffuser) accepted for Oral Presentation at ICANN 2025 (CCF-C).
- Two papers (LWSpace, EnergyPatchTST) accepted for Oral Presentation at ICIC 2025 (CCF-C).
- The self-developed CCF-Online academic conference portal is now live.
- Awarded Second Prize in the 'E-Cloud Cup' Shanghai Collegiate Cloud Computing Application Contest.
- Awarded Gold Medal in the university-level China 'Internet+' Innovation Competition.
Selected Publications
* denotes equal contribution.
-
2026
SDFlow: Similarity-Driven Flow Matching for Time Series Generation
Conference on Neural Information Processing Systems (NeurIPS 2026) -
2026
Adaptive Prototypical Contrastive Learning for Time Series Clustering
ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD 2026) -
2026
ClusterPatchTST: Uncertainty-Aware Causal Clustering for Heterogeneous Time Series Forecasting
International Conference on Database Systems for Advanced Applications (DASFAA 2026) -
2025
EnergyPatchTST: Multi-scale Time Series Transformers with Uncertainty Estimation for Energy Forecasting
International Conference on Intelligent Computing (ICIC 2025)
Experience
-
Long Time-Series Generation
ResearchResearch Collaboration | Advisors: Prof. Peilin Zhao (SJTU) & Dr. Shibo Feng (NTU)
Oct 2025 – Sep 2026
- Co-led SDFlow (NeurIPS 2026) as first-listed co-first author: similarity-driven, non-autoregressive Flow Matching in a low-rank VQ latent space for long-horizon generation.
- Extending the line to multiscale discrete representations and long-sequence generation, covering method design, implementation, and evaluation.
-
LLM-Based Multimodal Time-Series Alignment
ResearchResearch Collaboration | Advisors: Prof. Xiaofan Li (Nanjing University) & Dr. Jingge Xiao (Leibniz University Hannover)
Feb 2026 – Sep 2026
- Co-leading work on text-conditioned time-series generation: semantic prototype retrieval, latent-space localization, and residual Flow Matching (manuscript under review).
-
Pattern Recognition & Machine Learning
ResearchResearch Study | Mentor: Prof. Yuchun Fang (Shanghai University)
Oct 2025 – Sep 2026
- Studying pattern recognition and machine learning, with a focus on sign-language recognition, sequence modeling, and transferring time-series methods.
-
Time-Series Forecasting & Deep Learning
ResearchResearch Study | Mentor: Prof. Fenglei Yang (Shanghai University)
Jan 2024 – Aug 2024
- Worked on multiscale forecasting with SSM and Transformer architectures, building the modeling, experimentation, and writing foundation for later work such as EnergyPatchTST (ICIC 2025).
-
Cultural Heritage Data Management
ResearchUndergraduate Researcher | Key Laboratory of Silicate Cultural Heritage Protection, Shanghai University | Advisor: Prof. Shunbo Hu
Mar 2024 – Sep 2026
- Developed a specialized database using MySQL and Streamlit for managing cultural heritage data (Buddhist head sculptures).
- Participated in data analysis and management tasks related to cultural relic preservation.
-
LLM Capability Enhancement & Alignment
Part-timeByteDance | Xpert Program
Aug 2025 – Present
- Contribute to supervised fine-tuning and model evaluation for reasoning, search, tool use, and safety alignment.
Awards & Honors
- Baosteel ScholarshipShanghai University
- National ScholarshipMinistry of Education of China
- ASC Student Supercomputer ChallengeInternational First Prize
- Xiaomi ScholarshipXiaomi
- Baidu Star Programming ContestFinalist
- Special Class Scholarship for Academic ExcellenceShanghai University (Top-tier)
- Shanghai ScholarshipShanghai
- Baidu Star Programming Contest - Talent TrackNational First Prize
- MaTiBei Collegiate Programming ContestNational First Prize
- Information Security and Countermeasures ContestNational Second Prize
- 'Jie-Yi' Student Leadership ScholarshipSchool of Computer Engineering and Science
- 'E-Cloud Cup' Shanghai Collegiate Cloud Computing Application ContestSecond Prize
- Special Class Scholarship for Academic ExcellenceShanghai University (Top-tier)
- Undergraduate Academic ForumSecond Prize
- China 'Internet+' Innovation CompetitionGold Medal (University Level)
Projects & Systems
-
'CCF-Online' Academic Conference Portal
Independently developed and deployed an online portal that scrapes and integrates data from multiple sources to provide researchers with efficient access to academic conference information.
-
SHU Prophet: An Intelligent Time-Series Forecasting & Decision Platform
Engineered 'SHU Prophet,' a full-stack intelligent platform for advanced time-series analysis. This project integrates a suite of self-developed models with an interactive AI Agent (LangChain & LLM) to automate data analysis, generate insightful reports, and bridge the gap between complex algorithms and intuitive decision-making.
-
Cloud-based Automated Course Selection System
Led the development of a cloud-hosted system that automates the course selection process, featuring an LLM-based sentiment analysis module for course reviews. Award-winning project.
-
Cultural Heritage Database System
Developed a digital management platform for the University's Key Laboratory of Silicate Cultural Heritage, supporting multi-source data entry, retrieval, and visualization.
-
Smart Parking IoT Platform
Designed and built a full-stack IoT parking solution, enabling real-time parking availability queries, reservations, and payments. Awarded Gold Medal at a university innovation competition.
-
Research on Product Assembly Strategy Optimization
Led a research project applying Bayesian decision networks and genetic algorithms to optimize complex assembly strategies for products with variable components and processes.
Service & Talks
-
Academic Service
- Frontiers in Physics (Interdisciplinary Physics)Reviewer
- Conference on Neural Information Processing Systems (NeurIPS 2026)Ethics Reviewer (Invited)
- AAAI Conference on Artificial Intelligence (AAAI 2027)Program Committee Member / Reviewer (Invited)
- Neural Networks (Q1, CCF-B)Reviewer (Invited)
- Ain Shams Engineering Journal (Q1)Reviewer (Invited)
- International Conference on Artificial Neural Networks (ICANN 2026)Reviewer (Invited)
- AAAI Conference on Artificial Intelligence (AAAI 2026)Reviewer
- International Conference on Artificial Neural Networks (ICANN 2025)Reviewer (3 submissions)
-
Talks & Presentations
- Poster Presentation at KDD 2026Presented "APCL"
- Oral Presentation at DASFAA 2026Presented "ClusterPatchTST"
- Poster Presentation at ICASSP 2026Presented "ScatterFusion" and "AWGformer"
- Oral Presentation at ICANN 2025Presented "TimeFlowDiffuser" and "SWIFT"
- Oral Presentation at ICIC 2025Presented "LWSpace" and "EnergyPatchTST"
- Grade meetingInvited Speaker on academic and research journey
-
Leadership & University Service
- Head of Management Dept., School Youth League CommitteeSchool
- Class League Branch SecretaryShanghai University