ziyuye [at] uchicago.edu
I am a PhD student in Computer Science at The University of Chicago.
I work at Google DeepMind's Gemini team as a student researcher, previously a research intern.
I work on large model post-training for automated reasoning (e.g., theorem proving, competitive programming), with self-play methods in reinforcement learning.
I was the co-founder of Eigent AI, which backs up and is open-sourced at Camel AI.
I will join Google DeepMind as a Research Scientist late this summer.
[eva] Scalable Reinforcement Post-Training Beyond Static Human Prompts
Evolving Alignment via Asymmetric Self-Play
Ziyu Ye, Rishabh Agarwal, Tianqi Liu, Rishabh Joshi, Sarmishta Velury,
Quoc V. Le, Qijun Tan, Yuan Liu
An early version in Language Gamification Workshop at NeurIPS (spotlight), 2024.
[ paper ]
[ poster ]
[ slides ]
[RiR] Reasoning in Reasoning
A Hierarchical Framework for Neural Theorem Proving
Ziyu Ye, Jiacheng Chen, Jonathan Light, Yifei Wang, Jiankai Sun,
Mac Schwager, Philip Torr, Guohao Li, Yuxin Chen, Kaiyu Yang, Yisong Yue, Ziniu Hu
An early version in MATH-AI Workshop at NeurIPS, 2024.
[ paper ]
[ code ]
[ slides ]
Understanding the Role of Equivariance in Self-supervised Learning
Yifei Wang*, Kaiwen Hu*, Sharut Gupta, Ziyu Ye, Yisen Wang, Stefanie Jegelka
NeurIPS, 2024
[ paper ]
Don’t Be Pessimistic Too Early: Look K Steps Ahead (in Offline RL)
Chaoqi Wang, Ziyu Ye, Kevin Murphy, Yuxin Chen
AISTATS, 2023
[ paper ]
[ slides ]
Follow-ups Also Matter: Improving Contextual Bandits via Post-serving Contexts
Chaoqi Wang, Ziyu Ye, Zhe Feng, Ashwinkumar Badanidiyuru, Haifeng Xu
NeurIPS, 2023 (spotlight)
[ paper ]
[ slides ]
Provably Efficient Quantum Algorithms for Large-Scale Machine Learning Models
Junyu Liu, Minzhao Liu, Jin-Peng Liu, Ziyu Ye, Yuri Alexeev, Jens Eisert, Liang Jiang
Nature Communications, 2023
[ paper ]
[ code ]
Generalization and Memorization in Sparse Neural Networks
Ziyu Ye, Chaoqi Wang, Zixin Ding, Yuxin Chen
ICML Sparsity in Neural Networks Workshop, 2022
[ draft (unfinished) ]
[ blog ]
[ code ]
Understanding the Effect of Bias in Deep Anomaly Detection
Ziyu Ye, Yuxin Chen, Heather Zheng
IJCAI, 2021
[ paper ]
[ slides ]
[ code ]
I was formally trained to be an economist. My advisor was Prof. Yusen Kwoh, who was advised by Nobel Laureate Prof. Gary Becker.
I occasionally archive my grandfather's unpublished writings at this blog (e.g., Dream Thoughts). He has burned most of them during the cultural revolution.