About Me

I am a first-year Ph.D. student in the Department of Computer Science and Technology at Tsinghua University, fortunate to be advised by Prof. Wenguang Chen in the PACMAN Lab. My research focuses on Efficient Scaling of Large Language Models, with particular interest in pretraining Scaling Laws, to understand and optimize the relationship between model performance and training dynamics. I am fortunate to collaborate with Kaifeng Lyu and Shengding Hu.

I received my B.Eng. in Computer Science and Technology in 2024 from Tsinghua University, where I was a member of the Yao Class, headed by Prof. Andrew Chi-Chih Yao. During my undergraduate study, I was fortunately advised by Li Yi, Hao Su, and Wenguang Chen.

Publications

1. A Multi-Power Law for Loss Curve Prediction Across Learning Rate Schedules [ICLR 2025]
   Kairong Luo*, Haodong Wen, Shengding Hu, Zhenbo Sun, Zhiyuan Liu, Maosong Sun†, Kaifeng Lyu†, Wenguang Chen†
   • arXiv