Xingyu Fu (εΊζ妀)
π I'm a research scientist at Meta Superintelligence Labs, working on AI agents.
Before Meta, I was a Postdoctoral Fellow at Princeton University's PLI, working with Zhuang Liu, Danqi Chen, and Sanjeev Arora, during 2025β2026. I earned my Ph.D. in Computer Science at the University of Pennsylvania advised by Prof. Dan Roth from 2020 to 2025. During my PhD, I interned at Microsoft and AWS AI Labs. I did my B.S. in Computer Science at UIUC from 2017 to 2020, where I was very fortunate to be advised by Prof. Jiawei Han and Prof. Jingbo Shang.
I'm currently interested in building capable and self-improving AI agents β including agent evaluation, scalable self-improvement, and designing benchmarks that reveal emergent agent abilities. My research has also focused heavily on generative multimodal models at the intersection of vision and language (e.g., multimodal LLMs, text-to-image/video generation, omni models), where I built better evaluations for emergent abilities and used synthetic data to design models that can better perceive and reason about the multimodal world. My PhD thesis is Bridging Perception and Reasoning in Multimodal Models.
Paper of the day
