Hello! Husky here! I'm a MSML student at CMU Locus Lab. Throughout my research journey, I've explored a variety of topics. Currently, my primary focus is on generative world modeling, where I have been fortunate to work closely with Professor Zhiting Hu and Professor Yilun Du on advancing the capabilities of world models. Previously, I collaborated with my amazing advisor Jiaqi W. Ma at the TRAIS Lab, focusing on dataset curation using LLM agents. I also gained valuable research experience in AI interpretability under Professor Quanshi Zhang's XAI Lab.

My primary research goal is to build autonomous AI agents with human-level reasoning, planning, and memory that can flexibly allocate computation under varying budget constraints and adapt to evolving environments through closed-loop interaction. Currently, I am particularly interested in:

• Scalable Reasoning - methods that improve both capability and efficiency.
• World Modeling - video generative models for long horizon prediction, control, and planning.