Summary
Chun-hsiao Yeh is a PhD candidate at UC Berkeley with eight years of research experience at the intersection of computer vision and foundation models, focusing on spatial reasoning and geometric grounding for Multimodal Large Language Models. She has translated these perception advances into controllable generative pipelines that decompose and execute complex, instruction-driven image editing and synthesis. Her internships at Adobe and Meta FAIR underscore a track record of industry-facing research, while earlier work spans self-supervised representation learning, anomaly detection for autonomous driving, and face anti-spoofing. Based in Berkeley, she combines rigorous academic training with practical system-building, and she often leverages geometric awareness to make multimodal models more physically grounded—an approach that helps bridge perception and precise, controllable generation.
8 years of coding experience
8 years of employment as a software developer
Doctor of Philosophy - PhD, Vision Science (Computational Vision Track), Doctor of Philosophy - PhD, Vision Science (Computational Vision Track) at University of California, Berkeley
Master of Science (MS), Engineering Science, Division of Computer Science, Master of Science (MS), Engineering Science, Division of Computer Science at National Taiwan University
Exchange Program, Electrical Engineering and Computer Science, Exchange Program, Electrical Engineering and Computer Science at California State University, Long Beach
Bachelor of Science (BS), Electrical Engineering, Bachelor of Science (BS), Electrical Engineering at National Taiwan Normal University
English, Chinese, Mandarin, Japanese