Yuanchen Bei
I am a Ph.D. Student at the University of Illinois Urbana-Champaign, advised by Prof. Hanghang Tong.
My research interest lies in human-centered AI. My long-term goal is to build personalized and interactive agents capable of omni-modal reasoning. Toward this goal, I currently focus on memory management, proactiveness, multimodal understanding, and long-horizon post‑training for personalized agents.
Open to collaboration — feel free to reach out if you share interests in any of the directions below!
Research Highlight
Selected publications, grouped by research direction. * Equal contribution. Full list on Google Scholar.
Agent Memory and Context Management
- ACL 2026110+ GitHub Stars3000+ HuggingFace Monthly Downloads
- EMNLP 2026
Agent Reasoning and Post-Training
- NeurIPS 2026
User Modeling, Personalization, and Recommendation
- WWW 2026Applied at Alibaba
- KDD 2025Applied at Alibaba
- WSDM 2025Top-1 Most Cited Paper
- WWW 2024Top-10 Influential PaperApplied at Alibaba
- NeurIPS 2024Spotlight, Top 2%
Selected Open-Source Projects
-
A unified toolkit for cold-start recommendation. Includes 20+ models with comprehensive evaluations across multiple datasets.
-
A curated reading list of papers on graph-based retrieval-augmented generation.
-
A curated reading list of papers on cold-start recommendation.
Honors and Awards
- Outstanding Reviewer KDD 2025, TNNLS
- National Scholarship ×2 Bachelor, Master
- First Class Scholarship ×2 Bachelor, Master
- Outstanding Graduate Master
- Outstanding Undergraduate Bachelor
Academic Services
- Conference Program Committee Member: KDD (2024–2027), ICLR (2025–2027), WWW (2026–2027), WSDM (2026–2027), AAAI 2027, ICML 2026, NeurIPS 2026, ARR 2026, MM (2025–2026), CIKM (2025–2026), SIGIR 2025, SIGIR-AP 2025, ECAI 2024.
- Journal Reviewer: TMLR, IEEE TKDE, IEEE TNNLS, IEEE TII, IEEE TIP, ACM CSUR, ACM TKDD.
Misc
In my free time, I enjoy learning about finance and the global economy, and I highly recommend 小Lin说 for its engaging and accessible explainer videos. I’m also a movie lover (especially mysteries), a fan of Jisoo, and a dog lover.
Languages: Chinese (native), English (fluent), and Cantonese (proficient listening).