I am a Machine Learning Ph.D. student in the School of Electrical and Computer Engineering at the Georgia Institute of Technology, co-advised by Dr. Sara Fridovich-Keil in the VOILA Lab and Dr. Glen Chou in the Trustworthy Robotics Lab.

Research

My research lies at the intersection of generative modeling, robotics, and control theory. I study how pretrained generative models—particularly diffusion and flow models—can be made more controllable by viewing their sampling trajectories and internal representations as dynamical systems. My work explores both inference-time control, including activation-space steering with reduced-order dynamics and optimal control, and stochastic optimal control formulations for controllable and constrained generation.

I am particularly interested in extending these ideas to world models and world-action models for embodied AI. Here, my goal is to understand how generative models of physical dynamics can support robust prediction, planning, and action generation, and how they can be steered to remain reliable under distribution shift without requiring full retraining.

Publications

See the Publications page for the full list.