I’m Shanglin (Jason) Wu, a second-year Ph.D. student in Computer Science and Informatics at Emory University, where I am advised by Dr. Kai Shu. I received my Bachelor’s degree in Artificial Intelligence from Yuanpei College, Peking University in 2025.
My research focuses on understanding the fundamental principles that govern multi-agent AI systems, particularly how autonomous agents collaborate, learn, and adapt through interaction. I am especially interested in how collective behaviors emerge from individual agents, how agents learn continually from experience, and how coordination failures and emergent risks arise as these systems scale. My goal is to develop a deeper understanding of the mechanisms underlying learning, collaboration, and safety in increasingly autonomous multi-agent systems.
News
- 2026.06: 🎉🎉 Our paper Rethinking Memory Mechanisms of Foundation Agents in the Second Half: A Survey is accepted by TMLR!
- 2026.06: 🎉🎉 Our paper Memory in LLM-based Multi-agent Systems: Mechanisms, Challenges, and Collective Intelligence received Best Survey Paper Award 🏆 from PAKDD 2026!
- 2026.05: Honored to receive the Excellence in Teaching Assistance Commendation for year 2025–26 from the Department of Computer Science!
- 2026.04: Excited to join Cisco Research as AI/Intelligent Systems PhD Intern for Summer 2026!
- 2026.02: 🎉🎉 Our paper Memory in LLM-based Multi-agent Systems: Mechanisms, Challenges, and Collective Intelligence is accepted by PAKDD 2026!
- 2025.07: 🎉🎉 Completed my internship in Microsft Research Asia Alumni!
Publications
 | Memory in LLM-based Multi-agent Systems: Mechanisms, Challenges, and Collective Intelligence Shanglin Wu, Kai Shu Proceedings of the 30th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD 2026) |
 | Rethinking Memory Mechanisms of Foundation Agents in the Second Half: A Survey Wei-Chieh Huang, Weizhi Zhang, Yueqing Liang, ..., Shanglin Wu,..., Kai Shu Transactions on Machine Learning Research (TMLR). Core contributor. |
 | Improving Factuality in LLMs via Inference-Time Knowledge Graph Construction Shanglin Wu, Lihui Liu, Jinho D. Choi, Kai Shu |
 | Scaling Teams or Scaling Time? Memory Enabled Lifelong Learning in LLM Multi-Agent Systems Shanglin Wu, Yuyang Luo, Yueqing Liang, Kaiwen Shi, Yanfang Ye, Ali Payani, Kai Shu |
Research Experience
- Cisco Research, May 2026 - August 2026
- Microsoft Research Asia Alumni, Beijing, China, March 2025 - July 2025
Education
- 2025 - Now: Ph.D., Computer Science and Informatics. Emory University.
- 2021 - 2025: B.s., Artificial Intelligence. Peking University
Honors & Awards
- Excellence in Teaching Assistance Commendation, Emory University, 2026
- Awarded by the Department of Computer Science for exceptional dependability, proactivity, and initiative.
Academic Service
- Reviewer/Sub-reviewer: ICLR{2026}, TACL{2026}, WWW{2026}, ACL ARR{2025}, IEEE CogMI{2025}, IEEE BigData{2026}, SIGIR{2026}, COLM{2026}, Neurips{2026}, SDM{2026}, EMNLP{2026}.