I. 研究方向

1. AI Agent：研究基于大语言模型的智能体系统，重点关注智能体在开放环境中的工具使用、环境交互、复杂任务推理、未来预测与自我进化能力。具体方向包括：

•未来预测智能体：构建能够对未来事件进行预测，并基于真实结果反馈持续评估和改进的智能体系统。

•自我进化智能体：研究智能体如何通过反馈、搜索、反思和反复修正不断提升任务求解、推理与决策能力。

•代码与工具使用智能体：研究面向复杂软件仓库、动态工具调用和自动化任务执行的智能体学习、训练与评估方法。

2. AI for Finance：研究人工智能技术在金融市场建模与量化投资研究中的应用，重点关注市场动态建模、Alpha 因子挖掘与量化研究流程自动化。具体方向包括：

•市场微观结构世界模型：基于高频交易、订单流和市场微观结构数据刻画市场动态，为预测、模拟和交易决策提供支持。

•Alpha 因子自动挖掘：结合大语言模型、搜索、辩论和统计检验等方法，自动生成、解释和评估量化投资因子。

•自主量化研究智能体：构建支持研究假设生成、因子

II. 教育经历

2019.9 - 2025.6 清华大学 交叉信息研究院 计算机科学与技术 博士

2015.9 - 2019.6 北京航空航天大学 物理学院 应用物理学 学士

III. 工作经历

2025.7-至今 北京中关村学院 导师；中关村人工智能研究院 研究员

IV. 代表性学术论文

[1] Yunhe Han, Yunqi Gao, Bing Hu, Mahdi Boloursaz Mashhadi, Yitong Duan, Pei Xiao, Yanfeng Zhang. PipeSD: An Efficient Cloud-Edge Collaborative Pipeline Inference Framework with Speculative Decoding. ICML 2026. 

[2] Zhixin Han, Yanzhi Zhang, Chuyang Wei, Maohang Gao, Xiawei Yue, Kefei Chen, Yu Zhuang, Haoxiang Guan, Jiyan He, Jian Li, Yitong Duan, Yu Shi, Mengting Hu, Shuxin Zheng. FutureWorld: A Live Reinforcement Learning Environment for Predictive Agents with Real-World Outcome Rewards. arXiv, 2026. 

[3] Chuyang Wei, Maohang Gao, Zhixin Han, Kefei Chen, Yu Zhuang, Haoxiang Guan, Yanzhi Zhang, Yilin Cheng, Xiren Zhou, Huanhuan Chen, Jian Li, Jiyan He, Yu Shi, Yitong Duan, Shuxin Zheng. Harnessing Pre-Resolution Signals for Future Prediction Agents. arXiv, 2026. 

[4] Yanzhi Zhang, Zhaoxi Zhang, Haoxiang Guan, Yilin Cheng, Yitong Duan, Chen Wang, Yue Wang, Shuxin Zheng, Jiyan He. No Free Lunch: Rethinking Internal Feedback for LLM Reasoning. arXiv, 2025.

[5] Zhaoxi Zhang, Yitong Duan, Yanzhi Zhang, Yiming Xu, Zhixiang Wang, Kun Liang, Yang Li, Jiahui Liang, Deguo Xia, Jizhou Huang, Jiyan He, Yunfang Wu. One Tool Is Enough: Reinforcement Learning for Repository-Level LLM Agents. ICML 2026.

[6] Yanzhi Zhang, Yitong Duan, Zhaoxi Zhang, Jiyan He, Shuxin Zheng. Population-Evolve: a Parallel Sampling and Evolutionary Method for LLM Math Reasoning. arXiv, 2025.

[7] Yitong Duan, Chuheng Zhang, Jian Li. FactorMAD: A Multi-Agent Debate Framework Based on Large Language Models for Interpretable Stock Alpha Factor Mining. ICAIF 2025.

[8] Y. Shi, Yitong Duan, Jian Li. Navigating the Alpha Jungle: An LLM-Powered MCTS Framework for Formulaic Alpha Factor Mining. AAAI 2026.

[9] Yitong Duan, Weiran Wang, Jian Li. FactorGCL: A Hypergraph-Based Factor Model with Temporal Residual Contrastive Learning for Stock Returns Prediction. AAAI 2025.

[10] C. Zhang, Yitong Duan, X. Chen, J. Chen, J. Li, L. Zhao. Towards Generalizable Reinforcement Learning for Trade Execution. IJCAI 2023.

[11] Yitong Duan, Lei Wang, Qizhong Zhang, Jian Li. FactorVAE: A Probabilistic Dynamic Factor Model Based on Variational Autoencoder for Predicting Cross-Sectional Stock Returns. AAAI 2022.