Jichu Li李继初

About Me

I’m Jichu Li, currently an undergraduate student in the Honors Talent Program in Statistics and Data Science at Renmin University of China. I'm also a visiting student at Toyota Technological Institute at Chicago, hosted by Prof. Zhiyuan Li, and in collaboration with Prof. Tianhao Wang at UCSD. During my research journey, I had the fortune to work with Prof. Difan Zou on learning theory at The University of Hong Kong and Prof. Feng Zhou on multimodal event sequence modeling at Renmin University of China.

Research Interests

My research interests lie at the intersection of machine learning theory and optimization, with a focus on theoretical understanding of modern optimization algorithms in deep learning, analyzing their convergence, implicit bias, and generalization. I am also interested in empirically understanding the training dynamics and learning behavior of LLMs/MLLMs, with prior experience in LLM-based approaches for modeling complex sequential and event-based data.

I am currently seeking PhD opportunities in CS/Statistics/IEOR starting Fall 2027. Please feel free to email me if you’d like to connect!

News

• [Dec. 2026] I will be attending NeurIPS 2026 in Sydney, Australia. Looking forward to meeting you and chatting about research!
• [Sep. 2026] Our paper on learning with synthetic data via SGD was accepted to NeurIPS 2026.
• [Jul. 2026] I attended ICML 2026 in Seoul, South Korea.
• [May. 2026] Our work on learning with synthetic data via SGD was accepted to ICML 2026 Workshop on High-dimensional Learning Dynamics (HiLD).
• [Apr. 2026] Our paper on the implicit bias of stochastic steepest descent was accepted to ICML 2026.
• [Jan. 2026] Two papers on Convergence analysis of mixed-precision training and Multimodal event sequences modeling were accepted to ICLR 2026.
• [Sep. 2025] My first research project DanmakuTPPBench was accepted to NeurIPS 2025.

Publications

* Equal contribution. Google Scholar

Education

Renmin University of China
B.S. in Statistics, School of Statistics
B.Econ. (Minor) in Fintech, Gaoli Academy, Gaoling School of Artificial Intelligence.
Sep. 2023 – Jun. 2027 (expected) · Beijing, China
GPA: 3.97/4.00; ranking: 2/160+.

Research Visiting Experience

• Toyota Technological Institute at Chicago, Visiting Student
  Jul. – Sept. 2026 · Chicago, USA
  Hosted by Prof. Zhiyuan Li (on-site).
• The University of Hong Kong, Research Intern
  Jul. – Sept. 2025 · Hong Kong SAR, China
  Hosted by Prof. Difan Zou (on-site).

Awards & Honors

• [Sept. 2026] NeurIPS Travel Award, Neural Information Processing Systems Foundation.
• [Sept. 2026] Outstanding Student Award, Renmin University of China.
• [Sept. 2026] Academic Excellent Scholarship, School of Statistics, Renmin University of China. ￥8,000.
• [Sept. 2024 & 2025] National Scholarship (twice), highest scholarship for Chinese university students. ￥20,000.
• [Jul. 2025] HKU Summer Research Stipend, The University of Hong Kong. HK$19,600.
• [May 2025] Honorable Mention, Mathematical Contest in Modeling and Interdisciplinary Contest in Modeling.
• [Dec. 2024] Provincial First Prize, Mathematics Competition of Chinese College Students (CMC).
• [Nov. 2024] Provincial First Prize, Contemporary Undergraduate Mathematical Contest in Modeling.

Teaching

• Teaching Assistant, Optimization Methods, Fall 2026.

Services

Conference Reviewer

• Neural Information Processing Systems (NeurIPS) 2026
• International Conference on Machine Learning (ICML) 2026
• International Conference on Learning Representations (ICLR) 2027

Instructor

• Statistical Investigation Association of Renmin University of China, 2024 – present. Taught Python programming and wrote tutorial articles for student groups at Renmin University of China.

Blog

• Geometry Comparison in Gaussian Associative Memory, Oct. 2026.
• Can Next-Token Prediction Data Alone Favor Spectral Optimization? A Minimal Model Study, Oct. 2026.

Powered by Minimal Light theme.