Xutong Liu
About
Degrees
Working Experience
Postdoc, Electrical and Computer Engineering, Carnegie Mellon University, 2024 - 2025
Visiting Postdoc, CICS, University of Massachusetts Amherst, 2023 - 2024
Postdoc, Computer Science and Engineering, Chinese University of Hong Kong, 2022 - 2024
My Research
Dr. Xutong Liu’s research focuses on building structure-aware online learning and reinforcement learning (RL) algorithms that exploit the underlying action, feedback, and agent structures (e.g., smoothness, sparsity, clustering) to efficiently make optimal decisions (e.g., resource allocation, scheduling) in networked systems. He places a special emphasis on theory-grounded data-efficiency, scalability, and robustness guarantees for these RL algorithms, while also ensuring they can be readily applicable to real-world decision-making problems in edge-based multimedia systems, conversational recommendation systems, and cost-effective LLM serving systems. He was awarded the RGC Postdoctoral Fellowship, with his research recognized as the best paper runner-up at ACM SIGMETRICS 2025 and the long oral at ICML 2021. For more details, check my personal website.
The LEAD Research Lab
At UW, I am building the Learning, Evaluation, and Advanced Decision-making (LEAD) research lab and actively recruiting Ph.D. students in Fall 2026. Our lab also plans to recruit graduate/undergraduate interns, where students will have opportunities to collaborate with leading researchers from UW, CMU, UMass Amherst, Microsoft Research, and Adobe, etc. Check group information for details if you are interested in joining us.
Professional Services
Publicity Chair, the International European Conference on Parallel and Distributed Computing (Euro-Par) 2026
Technical Program Committee, ACM SIGMETRICS 2026
Technical Program Committee, ACM e-Energy 2026
Co-Organizer, Workshop on Learning-augmented Algorithms: Theory and Applications, SIGMETRICS 2025
Recent Publications
- [Preprint] Faster, Smaller, and Smarter: Task-Aware Expert Merging for Online MoE Inference
Ziyi Han, Xutong Liu, Ruiting Zhou, Xiangxiang Dai, John C.S. Lui.
[arXiv] - [SIGMETRICS '26] Heterogeneous Multi-agent Multi-armed Bandits on Stochastic Block Models
Mengfan Xu, Liren Shan, Fatemeh Ghaffari, Xuchuang Wang, Xutong Liu, and Mohammad Hajiesmaili.
Accepted by ACM International Conference on Measurement and Modeling of Computer Systems (SIGMETRICS), 2026 (23/111 = 20.7%).
[PDF] - [NeurIPS '25] Learning Across the Gap: Hybrid Multi-armed Bandits with Heterogeneous Offline and Online Data
Qijia He, Minghan Wang, Xutong Liu, Zhiyong Wang, Fang Kong.
The Thirty-nineth Conference on Neural Information Processing Systems (NeurIPS), 2025. (5290/21575=24.5%). - [Preprint] Semantic Caching for Low-Cost LLM Serving: From Offline Learning to Online Adaptation
Xutong Liu, Baran Atalar, Xiangxiang Dai, Jinhang Zuo, Siwei Wang, John C.S. Lui, Wei Chen, Carlee Joe-Wong.
[arXiv] - [SIGMETRICS '25, 🏆Best Paper Runner-Up] Combinatorial Logistic Bandits
Xutong Liu, Xiangxiang Dai, Xuchuang Wang, Mohammad Hajiesmaili, John C.S. Lui.
ACM International Conference on Measurement and Modeling of Computer Systems (SIGMETRICS), 2025 (5/223 = 2.3%).
Selected as Best Paper Runner-Up at SIGMETRICS 2025.
[arXiv] [code] [slide]