Zihao He
I'm a Research Scientist at Meta on the Core Ads Growth (CAG, formerly Monetization GenAI) team, where I build reward models, RLHF pipelines, and multimodal LLM-as-a-judge evaluation systems for image and video generation. My research interests include reinforcement learning from human feedback (RLHF), reward modeling, LLM alignment & evaluation, and vision-language models.
I obtained my PhD in Computer Science from USC, advised by Prof. Kristina Lerman in the SEA Lab. Before that, I spent a year as a master's student in Computer Technologies at Tsinghua University. I received my undergraduate degree in Communication Engineering from Beijing University of Posts and Telecommunications. I have also interned at TikTok, Amazon, and DiDi Global.