I am Yansheng Mao, an undergraduate student majoring in Computer Science at Peking University, expecting to graduate in 2026. After that, I will continue my studies at Peking University as a Ph.D. student in Artificial Intelligence. Currently, I am a member of Mu Lab, where I am fortunate to be advised by Prof. Muhan Zhang. My research interests broadly span the field of Large Language Models (LLMs), with a primary focus on model pre-training and continual learning. In my recent work, I have been investigating methods for enhancing the capability of LLMs to process and reason over long texts.

More generally, I am interested in advancing the foundations of LLMs and exploring how training paradigms can be improved to achieve more efficient adaptation and sustained performance over time.