Hongjia (Alex) Huang

My name is 黄泓嘉 (Hongjia Huang), also known as Alex. I am a master’s student in the MSR program at Carnegie Mellon’s Robotics Institute. I completed my B.S. in Computer Science and Mathematics at New York University Shanghai, graduating summa cum laude (GPA 3.93/4.0). I am broadly interested in building AI systems that are physically grounded — models that understand, simulate, and reason about the world as it actually works, from molecular dynamics to robot manipulation.

My research sits at the intersection of robotics, computer vision, and physics-grounded video generation. I am particularly drawn to questions about how physical constraints — ranging from atomic-scale interactions to macroscopic motion — can serve as inductive biases for learning representations that are stable, interpretable, and generalizable.

At NYU Shanghai I worked with Professor Shengjie Wang and Professor Tianyi Zhou on efficient training of graph neural networks for molecular force field prediction, using submodular data selection and geometry-aware latent representations. I continue to work with them on video-diffusion world-action models, asking what training-time auxiliary structure buys a frozen video-diffusion policy — particularly for spatial out-of-distribution generalization. At UMD I spent the summer of 2025 with Professor Furong Huang on TraceGen, which models the world in a compact 3D trace-space rather than in pixels so that a policy can learn from cross-embodiment, cross-environment video. I now work with Professor Huang and Professor Zhou on physics-informed vision-language-action models, integrating trajectory planning into action prediction.

My interest in physics traces back to competing in the Chinese Physics Olympiad (CPhO) in high school, which first made me think seriously about how to model the real world computationally.