Paper 'Theoretical limitations of multi-layer Transformer' published in FOCS 2025, authors: Lijie Chen, Binghui Peng, Hongxun Wu.
Paper 'Self-Attention Networks Can Process Bounded Hierarchical Languages' published in ACL 2021, authors: Shunyu Yao, Binghui Peng, Christos Papadimitriou, Karthik Narasimhan.
Paper 'Fast swap regret minimization and applications to approximate correlated equilibria' published in STOC 2024, authors: Binghui Peng, Aviad Rubinstein.
Paper 'High dimensional online calibration in polynomial time' published in FOCS 2025, author: Binghui Peng.
Paper 'Complexity of Equilibria in First-Price Auctions under General Tie-Breaking Rules' published in STOC 2023, authors: Xi Chen, Binghui Peng.
Paper 'Memory-Query Tradeoffs for Randomized Convex Optimization' published in FOCS 2023, authors: Xi Chen, Binghui Peng.
Paper 'Near Optimal Memory-Regret Tradeoff for Online Learning' published in FOCS 2023, authors: Binghui Peng, Aviad Rubinstein.
Paper 'Memory Bounds for Continual Learning' published in FOCS 2022, authors: Xi Chen, Christos Papadimitriou, Binghui Peng.
Paper 'The Complexity of Dynamic Least-Squares Regression' published in FOCS 2023, authors: Xi Chen, Binghui Peng.
Research Experience
Currently a visiting faculty researcher at Google Research, NYC. Previously, a postdoctoral research fellow at Stanford University (supervised by Aviad Rubinstein and Amin Saberi) and Simons Institute, UC Berkeley.
Education
Ph.D. from Columbia University, advised by Xi Chen and Christos Papadimitriou; Bachelor's degree from Yao Class at Tsinghua University.
Background
Research interests include machine learning theory, game theory, and theoretical computer science. Recently, particularly interested in the theory for LLM (Large Language Models).
Miscellany
Will join the computer science department of the University of Maryland starting from Jan 2026.