Xihuai Wang's Page

Shanghai, China

Xihuai Wang is currently a Ph.D. candidate at Shanghai Jiao Tong University, supervised by Prof. Weinan Zhang and Prof. Ying Wen. He is a member of SJTU-Apex Group and a member of SJTU-MARL Group leaded by Prof. Ying Wen, and is also selected into Wen-Tsun Wu AI Honorary Doctoral Class in 2020. Xihuai earned his B.Eng. in Computer Science and Technology at School of Computer Science and Engineering, Sun Yat-sen University in 2020.

Xihuai’s research interests include Decision-making and Multi-agent System. Specifically, he is now focusing on

Multi-agent Decision-making in Cooperative Scenarios, especially
- Efficiency of Cooperative Multi-agent Reinforcement Learning;
- Zero-shot Generalization Ability in Cooperative Multi-agent Systems.
Large Language Models for Decision-making, specifically human-ai collaboration.

news

Sep 26, 2024	Our work about zero-shot coordination evaluation ZSC-Eval is accepted by NeurIPS 2024 Dataset and Benchmark Track!
Aug 8, 2023	Give a talk about cooperative multi-agent reinforcement learning (Coordinate Agents vis Policy Optimization) at RLChina.
Mar 25, 2023	Our work about policy optimization in cooperative multi-agent scenarios Order Matters: Agent-by-agent Policy Optimization is accepted by ICLR 2023!

selected publications

Language Agent
Leveraging Dual Process Theory in Language Agent Framework for Real-time Simultaneous Human-AI Collaboration

Shao Zhang*, Xihuai Wang*, Wenhao Zhang, Chaoran Li, Junru Song, Tingyu Li, Lin Qiu, Xuezhi Cao, Xunliang Cai, Wen Yao, Weinan Zhang, Xinbing Wang, and Ying Wen

Preprint Under Review, 2025

Abs Bib PDF Code

Agents built on large language models (LLMs) have excelled in turn-by-turn human-AI collaboration but struggle with simultaneous tasks requiring real-time interaction. Latency issues and the challenge of inferring variable human strategies hinder their ability to make autonomous decisions without explicit instructions. Through experiments with current independent System 1 and System 2 methods, we validate the necessity of using Dual Process Theory (DPT) in real-time tasks. We propose DPT-Agent, a novel language agent framework that integrates System 1 and System 2 for efficient real-time simultaneous human-AI collaboration. DPT-Agent’s System 1 uses a Finite-state Machine (FSM) and code-as-policy for fast, intuitive, and controllable decision-making. DPT-Agent’s System 2 integrates Theory of Mind (ToM) and asynchronous reflection to infer human intentions and perform reasoning-based autonomous decisions. We demonstrate the effectiveness of DPT-Agent through further experiments with rule-based agents and human collaborators, showing significant improvements over mainstream LLM-based frameworks. To the best of our knowledge, DPT-Agent is the first language agent framework that achieves successful real-time simultaneous human-AI collaboration autonomously. Code of DPT-Agent can be found in https://github.com/sjtu-marl/DPT-Agent.
@article{zhang2025dpt, title = {Leveraging Dual Process Theory in Language Agent Framework for Real-time Simultaneous Human-AI Collaboration}, author = {Zhang*, Shao and Wang*, Xihuai and Zhang, Wenhao and Li, Chaoran and Song, Junru and Li, Tingyu and Qiu, Lin and Cao, Xuezhi and Cai, Xunliang and Yao, Wen and Zhang, Weinan and Wang, Xinbing and Wen, Ying}, year = {2025}, journal = {Preprint Under Review}, }
MARL Generalization

ZSC-Eval: An Evaluation Toolkit and Benchmark for Multi-agent Zero-shot Coordination

Xihuai Wang, Shao Zhang, Wenhao Zhang, Wentao Dong, Jingxiao Chen, Ying Wen, and Weinan Zhang

38th NeurIPS Dataset and Benchmark Track, 2024

PDF Code
MARL Efficiency

Order Matters: Agent-by-agent Policy Optimization

Xihuai Wang, Zheng Tian, Ziyu Wan, Ying Wen, Jun Wang, and Weinan Zhang

11th ICLR, 2023

PDF Code
MARL Efficiency

Model-based Multi-agent Policy Optimization with Adaptive Opponent-wise Rollouts

Weinan Zhang, Xihuai Wang, Jian Shen, and Ming Zhou

30th IJCAI, 2021

PDF Code