Bio
I am an AI Research Engineer at Duolingo. I received my Ph.D. from the Department of Electrical, Computer, and Systems Engineering at Rensselaer Polytechnic Institute (RPI), where I was advised by Prof. Ali Tajer. Before that, I obtained my B.S. in Statistics and B.E. in Computer Science and Technology from the University of Science and Technology of China in 2020.
My research focuses on causal machine learning and sequential decision-making, with an emphasis on causal bandits (see this GitHub repository for a brief overview) and causal analysis of large language models (LLMs). I am also interested in bandit applications in federated learning.
Experience
June 2026 - PresentAI Research Engineer II, Monetization Engine
May 2025 - Aug 2025AI Research Engineer Intern, Monetization Engine
June 2024 - Aug 2024Research Intern, Search and Recommendation Science
IBM Thomas J. Watson Research Center
May 2023 - Aug 2023Research Extern, Trustworthy AI
Education
Rensselaer Polytechnic Institute
2021 - 2026Ph.D. in Electrical, Computer, and Systems Engineering
Advisor: Prof. Ali Tajer
Dissertation: Causal Bandits with Soft Intervention: A Unified Framework
University of Science and Technology of China
2016 - 2020B.S. in Statistics and B.E. in Computer Science and Technology
Publications(* equal contribution)
Causal Bayesian Optimization via Causal Bandits: Open Questions
Arpan Mukherjee, Zirui Yan, Ali Tajer
Proc. Conference on Uncertainty in Artificial Intelligence Workshop on Causality for Decision Making (UAI CDM), 2026.
Multi-component Causal Tracing in Large Language Models
Zirui Yan, Dennis Wei, Dmitriy A. Katz, Prasanna Sattigeri, Ali Tajer
Proc. Annual Meeting of the Association for Computational Linguistics (ACL), 2026. Oral
Reward-oriented Causal Representation Learning
Zirui Yan*, Emre Acartürk*, Ali Tajer
Proc. Neural Information Processing Systems (NeurIPS), 2025.
Linear Causal Bandits: Unknown Graph and Soft Interventions
Zirui Yan, Ali Tajer
Proc. Neural Information Processing Systems (NeurIPS), 2024.
Improved Bound for Robust Causal Bandits with Linear Models
Zirui Yan, Arpan Mukherjee, Burak Varıcı, Ali Tajer
Proc. IEEE International Symposium on Information Theory (ISIT), 2024.
Causal Bandits with General Causal Models and Interventions
Zirui Yan, Dennis Wei, Dmitriy A. Katz, Prasanna Sattigeri, Ali Tajer
Proc. International Conference on Artificial Intelligence and Statistics (AISTATS), 2024.
Robust Causal Bandits for Linear Models
Zirui Yan, Arpan Mukherjee, Burak Varıcı, Ali Tajer
IEEE Journal on Selected Areas in Information Theory (JSAIT), 2024.
Optimizing Parameter Mixing under Constrained Communications in Parallel Federated Learning
Xuezheng Liu*, Zirui Yan*, Yipeng Zhou , Di Wu , Xu Chen , Jessie Hui Wang (* equal contributor)
IEEE/ACM Trans. on Networking (TON), 2023.
Federated Multi-armed Bandit via Uncoordinated Exploration
Zirui Yan, Quan Xiao, Tianyi Chen, Ali Tajer
Proceedings of IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2022.
Image denoising via K-SVD with primal-dual active set algorithm
Quan Xiao, Canhong Wen, Zirui Yan
Proc. Winter Conference on Applications of Computer Vision (WACV), 2020.
Academic Service
- Reviewer of IEEE Journal on Selected Areas in Information Theory
- Reviewer of ICML 2025
- Reviewer of TMLR 2025
- Reviewer of ICLR 2025
- Reviewer of NeurIPS 2024 (top reviewer), 2025, 2026
- Reviewer of Transactions on Machine Learning Research
- Reviewer of IEEE Transactions on Signal Processing
- Reviewer of AISTATS 2024
- Reviewer of ICML 2023 Workshop on Structured Probabilistic Inference & Generative Modeling (SPIGM)