I am currently a senior researcher at Tencent. I obtained my Ph.D. (2021) and B.S. (2016) in Computer Science at Tsinghua University, advised by Guangwen Yang.

Previously I was a visiting researcher at Stanford, where I was fortunate to advised by Tengyu Ma. Before that, I spent three years of internship at Microsoft Research and worked closely with Lintao Zhang, Tao Qin and Li Zhao.

I am currently interested in research topics related to multimodal reasoning, agent, reinforcement learning, causal inference and their commercial applications. Email: lzcthu12[at]gmail.com

Publications & Preprints

($^*$indicates equal contribution, $^{\dagger}$indicates project lead)

Multimodal Reasoning & Agents

• UI-Voyager: A Self-Evolving GUI Agent Learning via Failed Experience
  Zichuan Lin*$^{\dagger}$, Feiyu Liu*, Yijun Yang*, Jiafei Lyu*, Yiming Gao*, Yicheng Liu*, Zhicong Lu, Yangbin Yu, Mingyu Yang, Junyou Li, Deheng Ye, Jie Jiang
  Technical Report, 2026 [Code] [Huggingface] (#2 Daily Paper)

• AdaptVision: Efficient Vision-Language Models via Adaptive Visual Acquisition
  Zichuan Lin*$^{\dagger}$, Yicheng Liu*, Yang Yang, Lvfang Tao, Deheng Ye
  CVPR 2026 [Code] [Project Page]

• ProAct: Agentic Lookahead in Interactive Environments
  Yangbin Yu, Mingyu Yang, Junyou Li, Yiming Gao, Feiyu Liu, Yijun Yang, Zichuan Lin, Jiafei Lyu, Yicheng Liu, Zhicong Lu, Deheng Ye, Jie Jiang
  Technical Report, 2026 [Code]

• CausalMACE: Causality Empowered Multi-Agents in Minecraft Cooperative Tasks
  Qi Chai, Zhang Zheng, Junlong Ren, Deheng Ye, Zichuan Lin, Hao Wang
  EMNLP 2025

• SeeNav-Agent: Enhancing Vision-Language Navigation with Visual Prompt and Step-Level Policy Optimization
  Zhengcheng Wang*, Zichuan Lin*, Yijun Yang, Haobo Fu, Deheng Ye
  Preprint, 2025 [Code]

• EntroPIC: Towards Stable Long-Term Training of LLMs via Entropy Stabilization with Proportional-Integral Control
  Kai Yang, Xin Xu, Yangkun Chen, Weijie Liu, Jiafei Lyu, Zichuan Lin, Deheng Ye, Saiyong Yang
  Preprint, 2025 [Code]

Reinforcement Learning

• Multi-agent In-context Coordination via Decentralized Memory Retrieval
  Tao Jiang, Zichuan Lin$^{\dagger}$, Lihe Li, Yi-Chen Li, Cong Guan, Lei Yuan, Zongzhang Zhang, Yang Yu, Deheng Ye
  AAAI 2026 (Oral) [Code]

• Revisiting Discrete Soft Actor-Critic
  Haibin Zhou, Tong Wei, Zichuan Lin, Junyou Li, Deheng Ye, Qiang Fu, Wei Yang
  TMLR 2024

• CurrMask: Learning Versatile Skills with Automatic Masking Curricula
  Zhihui Xie, Yao Tang, Zichuan Lin, Deheng Ye, Shuai Li
  NeurIPS 2024

• Replay-enhanced Continual Reinforcement Learning
  Tiantian Zhang, Kevin Z. Shen, Zichuan Lin, Bo Yuan, Xueqian Wang, Xiu Li, Deheng Ye
  TMLR 2023

• A Survey on Transformers in Reinforcement Learning
  Wenzhe Li*, Hao Luo*, Zichuan Lin*, Chongjie Zhang, Zongqing Lu, Deheng Ye
  TMLR 2023

• Future-conditioned Unsupervised Pretraining for Decision Transformer
  Zhihui Xie, Zichuan Lin, Deheng Ye, Qiang Fu, Wei Yang, Shuai Li
  ICML 2023

• Dynamics-Adaptive Continual Reinforcement Learning via Progressive Contextualization
  Tiantian Zhang, Zichuan Lin, Yuxing Wang, Deheng Ye, Qiang Fu, Wei Yang, Xueqian Wang, Bin Liang, Bo Yuan, Xiu Li
  TNNLS 2023

• JueWu-MC: Playing Minecraft with Sample-efficient Hierarchical Reinforcement Learning
  Zichuan Lin*, Junyou Li*, Jianing Shi*, Deheng Ye, Qiang Fu, Wei Yang
  IJCAI 2022 (Long Oral top3%) (The champion solution of NeurIPS 2021 MineRL research competition)

• Model-based Adversarial Meta-Reinforcement Learning [code]
  Zichuan Lin, Garrett Thomas, Guangwen Yang, Tengyu Ma
  NeurIPS 2020

• RD$^2$: Reward Decomposition with Representation Decomposition
  Zichuan Lin*, Derek Yang*, Li Zhao, Tao Qin, Guangwen Yang, Tie-yan Liu
  NeurIPS 2020

• Episodic Reinforcement Learning with Associative Memory
  Guangxiang Zhu*, Zichuan Lin*, Guangwen Yang, and Chongjie Zhang
  ICLR 2020

• Object-Oriented Dynamics Learning through Multi-Level Abstraction
  Guangxiang Zhu*, Jianhao Wang*, Zhizhou Ren*, Zichuan Lin and Chongjie Zhang
  AAAI 2020

• Distributional Reward Decomposition for Reinforcement Learning
  Zichuan Lin, Li Zhao, Derek Yang, Tao Qin, Guangwen Yang, and Tie-yan Liu
  NeurIPS 2019

• Fully Parameterized Quantile Function for Distributional Reinforcement Learning [code]
  Derek Yang, Li Zhao, Zichuan Lin, Jiang Bian, Tao Qin, and Tie-yan Liu
  NeurIPS 2019

• Episodic Memory Deep Q-Networks [code]
  Zichuan Lin, Tianqi Zhao, Guangwen Yang, and Lintao Zhang
  IJCAI 2018

• Sample Dropout: A Simple yet Effective Variance Reduction Technique in Deep Policy Optimization
  Zichuan Lin, Xiapeng Wu, Mingfei Sun, Deheng Ye, Qiang Fu, Wei Yang, Wei Liu
  arXiv:2302.02299, 2023

• Pretraining in Deep Reinforcement Learning: A Survey
  Zhihui Xie, Zichuan Lin, Junyou Li, Shuai Li, Deheng Ye
  arXiv:2211.03959, 2022

• Joint System-Wise Optimization for Pipeline Goal-Oriented Dialog System
  Zichuan Lin, Jing Huang, Bowen Zhou, Xiaodong He, Tengyu Ma
  arXiv:2106.04835, 2021

Causal Inference

• PIPCFR: Pseudo-outcome Imputation with Post-treatment Variables for Individual Treatment Effect Estimation
  Zichuan Lin*, Xiaokai Huang*, Jiate Liu, Yuxuan Han, Jia Chen, Xiapeng Wu, Deheng Ye
  arXiv:2512.18737, 2025 [Code]

Awards & Honors

• 北京市优秀博士生, 2021.
• 清华国家奖学金 (Top 1%) , 2018 & 2020.
• 清华大学年度人物提名，每年20名, 2019.
• 清华微信公众号采访, 2018.
• 清华大学十佳优秀运动员，每年10名, 2017.
• 清华计算机系本科优秀毕业论文, 2016.
• 首都高校乒乓球比赛男子单打四连冠, 2013-2016.
• 全国中学生乒乓球锦标赛男子双打第三名, 2012.
• 乒乓球国家一级运动员, 2006 & 2012.
• 世界中学生乒乓球锦标赛团体冠军, 2011.