Evan Z. Liu
Researcher @ Google DeepMind
Hi! I'm a researcher at Google DeepMind focused on reasoning and post-training research. I co-led the development of the recipe that became Gemini 3 DeepThink and contributed to Gemini 3.0 and Gemini 3.1. I also worked on the reasoning model that won a gold medal at ICPC World Finals.
Previously, I completed my PhD at Stanford University, advised by Chelsea Finn. My thesis focused on building pratical agents that can quickly solve new tasks by leveraging prior experience on related tasks via meta-reinforcement learning.
I spend most of my free time trying not to pull too hard on plastic or real rock (rock climbing). I'd love to climb with you!
Selected Publications (See all)
-
Meta-Reinforcement Learning: Algorithms and ApplicationsEvan Zheran Liu
PhD Dissertation, 2023 -
Giving Feedback on Interactive Programs with Meta-ExplorationEvan Zheran Liu*, Moritz Stephan*, Allen Nie, Emma Brunskill, Chris Piech, Chelsea Finn
Neural Information Processing Systems (NeurIPS), 2022 (oral) -
Decoupling Exploration and Exploitation in Meta-Reinforcement Learning without SacrificesEvan Zheran Liu, Aditi Raghunathan, Percy Liang, Chelsea Finn
International Conference on Machine Learning (ICML), 2021 -
An Imitation Learning Approach for Cache ReplacementEvan Zheran Liu, Milad Hashemi, Kevin Swersky, Parthasarathy Ranganathan, Junwhan Ahn
International Conference on Machine Learning (ICML), 2020
-
Learning Options via CompressionYiding Jiang*, Evan Zheran Liu*, Benjamin Eysenbach, Zico Kolter, Chelsea Finn
Neural Information Processing Systems (NeurIPS), 2022 -
Just Train Twice: Improving Group Robustness without Training Group InformationEvan Zheran Liu*, Behzad Haghgoo*, Annie S. Chen*, Aditi Raghunathan, Pang Wei Koh, Shiori Sagawa, Percy Liang, Chelsea Finn
International Conference on Machine Learning (ICML), 2021 (long talk)
Contact
You can reach me at [my first name][my last name]@cs.stanford.edu
Last updated: March 13, 2026