Hi! I am Kai Yang (杨铠), a final-year undergraduate student in Turing Class, School of EECS, Peking University. I’m fortunate to be advised by Prof. Liwei Wang and Prof. Di He. During the summer of 2024, I worked as an student intern advised by Prof. Pradeep Ravikumar in Carnegie Mellon University.

My research interest lies in machine learning theory and algorithms. My work applies mathematical tools, aiming to provide theoretical insights into the capabilities and limitations of contemporary models and algorithms, and possibly design new ones with provable guarantees. Here are some research topics I work on:

• The foundations of Transformers and large language models in complex tasks.
• The mechanisms of large-scale pretraining algorithms and representation learning.

Always feel free to contact me through email if you are interested in my research or want to have a chat :)

🔥 News

• 2025.05:  🎉🎉 One paper Contextures: Representations from Contexts is accepted at ICML 2025!
• 2025.05:  🎉🎉 One paper How Numerical Precision Affects Arithmetical Reasoning Capabilities of LLMs is accepted at ACL 2025 Findings!
• 2025.04:  🎉🎉 I will join the ECE Department at the University of Michigan as a Ph.D. student in August, 2025!

🏫 Professional Services

Peer Review:

• ICLR: 2025.
• NeurIPS workshop: M3L 2024.
• ICML workshop: MOSS 2025.

Teaching Assistant:

• Peking University: Mathematical Foundations for the Information Age, Fall 2023 & Fall 2024.
• Peking University: Randomized Algorithms, Spring 2024.

🎖 Honors and Awards

• May Fourth Scholarship, The highest level scholarship at Peking University, 2023 - 2024.
• First Prize, 14th & 15th National University Mathematical Contest, 2022 & 2023.
• Silver Medal, 36th Chinese Mathematical Olympiad, 2020.