Lexin Zhou

I’m a CS PhD candidate at Princeton University, advised by Peter Henderson and closely collaborate with Tom Griffiths. I apply theories and experimental paradigms from cognitive science and psychometrics to evaluate frontier AI systems, value data recipes, and guide training decisions to improve real-world generalization and reliability. A few technical directions that I’m recently excited about:

• Robust evaluations that track intended behavior despite (intense) optimization pressure
• Translating large-scale evaluation insights into data valuation and synthetic environment generation
• Continual learning of AI agents

I’ve spent time in research/consultancy roles at Google DeepMind, MSR, OpenAI, Meta AI, and European Commission. My work has been featured in Nature, Financial Times, Microsoft Research, MIT Tech Review, Forbes, IEEE Spectrum, El País, New Scientists, QbitAI, IBM, among others. I am grateful to have my latest and ongoing projects being generously supported by OpenAI, Google DeepMind and MSR.

I’m always open to interesting conversations. Feel free to reach out via email or Twitter.