I am Yi Xie, a Ph.D. student in Computer Science & Engineering at the University of Arizona, advised by Prof. Bo Liu. Previously, I received my M.S. from Fudan University (advised by Prof. Zhongxue Gan) and dual B.S. degrees from Beijing University of Chemical Technology. I also work closely with Prof. Bo Han at Hong Kong Baptist University.

My research centers on building reliable and principled multi-agent LLM systems — from post-training with theoretical guarantees to trustworthy evaluation and code reasoning. I aim to develop LLM agents that can reason collaboratively, be rigorously evaluated, and solve complex real-world problems. My work spans three directions:

• Multi-Agent LLM Reasoning & Post-Training: How can multiple LLMs collaborate and improve with provable guarantees?
• Reliable Reasoning Evaluation: Where are the boundaries of LLM reasoning, and how do we measure them faithfully?
• Agentic Training: How can LLMs work with traceable, controllable reasoning in long trajectory for real world task?

I welcome collaborations in multi-agent LLM systems, reasoning evaluation, and code intelligence. Feel free to reach out via email.

E-mail: yix [at] arizona.edu

🔥 News

• 2026.03: “Modality Dominance-Aware Optimization for Embodied RGB–Infrared Perception” is accepted at ICME 2026 as [Spotlight]
• 2026.01: “SAT: Sequential Agent Tuning” is accepted at AAMAS 2026!
• 2025.09: Started my Ph.D. at the University of Arizona!
• 2025.05: “From Debate to Equilibrium” is accepted at ICML 2025!
• 2025.01: Two papers accepted at AAMAS 2025!

📝 Selected Publications

AAMAS 2026

SAT: Sequential Agent Tuning for Coordinator-Free Plug-and-Play Multi-LLM Training with Monotonic Improvement Guarantees

Yi Xie, Yangyang Xu, Fan Yi, Bo Liu

AAMAS 2026 [paper] [code]

ICML 2025

From Debate to Equilibrium: Belief-Driven Multi-Agent LLM Reasoning via Bayesian Nash Equilibrium

Yi Xie, Zhanke Zhou, Chentao Cao, Qiyu Niu, Tongliang Liu, Bo Han

ICML 2025 [paper] [code]

AAMAS 2025

ACORN: Acyclic Coordination with Reachability Network to Reduce Communication Redundancy in Multi-Agent Systems

Yi Xie, Ziqing Zhou, Chun Ouyang, Siao Liu, Zhile Zhao, Zhongxue Gan

AAMAS 2025 [paper] [code]

IEEE TIE

Improving Robotic Grasp Detection Under Sparse Annotations via Grasp Transformer with Pixel-Wise Contrastive Learning, Siao Liu, Yang Liu, Zhaoyu Chen, Ziqing Zhou, Zhile Zhao, Yi Xie, Wei Li, Zhongxue Gan.

IEEE Transactions on Industrial Electronics, 2025 [paper]

ICCV 2023

Improving Generalization in Visual Reinforcement Learning via Conflict-aware Gradient Agreement Augmentation, Siao Liu, Zhaoyu Chen, Yang Liu, Yuzheng Wang, Dingkang Yang, Zhile Zhao, Ziqing Zhou, Yi Xie, Wei Li, Wenqiang Zhang, Zhongxue Gan.

ICCV 2023 [paper]

💻 Service

• Conference Reviewer: NeurIPS 2024, 2025, 2026; ICLR 2025, 2026; ICML 2025, 2026; COLM 2025
• Journal Reviewer: KBS, IEEE TII, IEEE TNNLS