Jiarui Wang
I’m a Ph.D. student in Computer Science at the School of Computer Science, Shanghai Jiao Tong University (SJTU), advised by Prof. Zhouhan Lin in Language Understanding and Machine Intelligence Algorithms (LUMIA) Group. I received my B.S. in Computer Science from SJTU.
My research focuses on machine learning and large language models (LLMs), with particular interests in:
Next-Generation LLM Architectures — exploring new architectural paradigms that decouple knowledge storage from reasoning to improve knowledge acquisition, reasoning, and generalization.
Parametric Memory — developing parametric memory modules for LLMs that efficiently acquire, retain, and update knowledge.
I’m always open to discussing research, collaboration, and new ideas. Please feel free to reach out!
🔥 News
- 2026.07 🚀 Intern-S2-Preview-397B released — led the research and development of its memory architecture.
- 2026.07 🚀 MemSFT released — paper, model weights, and code are now open-sourced.
- 2026.07 🚀 Memory Decoder at Scale released.
- 2026.01 🎉 MLP Memory accepted to ICLR 2026
- 2025.10 🚀 MLP Memory released.
- 2025.09 🚀 Memory Decoder released.
- 2025.09 🎉 Memory Decoder accepted to NeurIPS 2025
- 2024.05 🏆 Our Robocup team SRC won Second Place in RoboCup China Open
📚 Publications
-
MemSFT: Mitigating Alignment Tax with an External Parametric Memory
Jiarui Wang*, Xiang Shi*, Jiaqi Cao, Rubin Wei, Xiquan Wang, Hao Sun, Jingzhi Wang, Zhiqi Yang, Qipeng Guo, Bowen Zhou, Zhouhan Lin
arXiv preprint (2026)
🔗 arXiv · GitHub · Hugging Face · * Equal Contribution -
Intern-S2-Preview: Scientific Agentic Foundation Model
Intern-S2-Preview Team, Shanghai AI Laboratory
Co-author; arXiv preprint (2026)
🔗 arXiv · GitHub · Hugging Face -
Memory Decoder at Scale: A Pretrained, Parametric Long-Term Memory
Rubin Wei, Jiaqi Cao, Jiarui Wang, Junming Zhang, Qipeng Guo, Bowen Zhou, Zhouhan Lin
arXiv preprint (2026)
🔗 arXiv · GitHub · Hugging Face -
Deep Fusion Memory: A Chunk-Level Parametric Memory for Large Language Models
Rubin Wei, Siyuan Sun, Hao Sun, Yuliang Liu, Xiang Shi, Jiarui Wang, Junming Zhang, Zhouhan Lin
arXiv preprint (2026)
🔗 arXiv -
HABIT-Bench: Evaluating Habit Induction from Longitudinal Weak Evidence in Agent Memory
Junming Zhang, Rubin Wei, Xu Cheng, Hao Sun, Jiarui Wang, Ruoxi Cheng, Xiquan Wang, Peisong Wu, Mingsong Chen, Zhouhan Lin
arXiv preprint (2026)
🔗 arXiv -
MLP Memory: A Retriever-Pretrained Memory for Large Language Models
Rubin Wei*, Jiaqi Cao*, Jiarui Wang, Jushi Kai, Qipeng Guo, Bowen Zhou, Zhouhan Lin
ICLR 2026
🔗 arXiv · GitHub · Hugging Face · * Equal Contribution -
Memory Decoder: A Pretrained, Plug-and-Play Memory for Large Language Models
Jiaqi Cao*, Jiarui Wang*, Rubin Wei, Qipeng Guo, Kai Chen, Bowen Zhou, Zhouhan Lin
NeurIPS 2025
🔗 arXiv · GitHub · Hugging Face · * Equal Contribution
🎓 Education
- 2025.09–Present, Ph.D. Student, School of Computer Science, Shanghai Jiao Tong University
- 2021.09–2025.06, B.E. in Computer Science, Shanghai Jiao Tong University
💼 Internships
- 2025.08–Present, Shanghai AI Lab, Shanghai, China
🏅 Honors and Awards
- Second Place, RoboCup China Open - Small Size League, 2024
- Third Prize, Prototype System Competition, CCF Chinasoft, 2023
- Provincial First Prize, Senior Group, National Olympiad in Informatics in Provinces (NOIP), 2018
