> Efficient AI · DLSys · Robotics

I am a master’s student in Computer Science at Carnegie Mellon University, advised by Prof. Chenyan Xiong (CMU), research scientist Han Cai (NVIDIA), and Prof. Song Han (MIT and NVIDIA). My main research interest is the co-design of algorithms and infrastructure for efficient training and inference of large models.

I also earned my B.S. in Computer Science (System and Network Concentration) with an additional concentration in Robotics from CMU. During my undergraduate studies, I was a member of the Catalyst Group, advised by Prof. Todd C. Mowry.

Efficient AI

My research is dedicated to enhancing the efficiency and long-context capacity of LLMs through co-optimization of algorithms and underlying parallel computer systems. Outside CMU, I work on efficient LLMs and am fortunate to be mentored by Prof. Song Han and Research Scientist Han Cai at MIT HAN Lab and NVIDIA.

Robotics

Before I started doing research on LLMs, I spent 1.5 years of my undergraduate studies conducting robotics research in DROP Lab at CMU, where I was mentored by Prof. Matthew Johnson-Roberson (now Dean of CCC at Vanderbilt University) and Prof. Weiming Zhi. I’ve worked with robot manipulators, quadrupeds, and humanoids, and my main focus in robotics was robotic perception for manipulation and robot foundation models.

Physics

I used to be a big fan of subatomic particles and quantum physics. I won awards in Physics Olympiad competitions and conducted physics research before beginning my undergraduate studies. Check out the bottom of the Publications & Projects page.

The enemy of memory isn’t time; it’s other memories.

Miscellaneous

I like playing and watching soccer. I played right back and on the wing for my school teams from middle school through high school. Since starting college, I’ve transitioned to the midfield and stay active by joining pickup games, intramural leagues, and the occasional community match. I also enjoy playing badminton recreationally.