👨‍🎓 About Me

Hi! I am a PhD student at BAIR, UC Berkeley, advised by Prof. Koushil Sreenath. Prior to that, I worked as a research intern at AIR-DREAM Lab of Institute for AI Industry Research (AIR), Tsinghua University, advised by Prof. Xianyuan Zhan. I obtained B.S. and M.S. in 2022 and 2025 from the Department of Automation, Tsinghua University. My research interest lies in reinforcement learning and cross-domain policy adaptation (i.e. sim-to-real) for embodied agents.

🔥 News

• [2025.12] One Paper (xTED) is accepted by AAMAS 2026!
• [2025.05] Invited talk about LBP (ICML 2025) at RLChina Academic Seminar!
• [2025.05] One paper (LBP) is accepted by ICML 2025!
• [2025.01] One paper (H2O+) is accepted by ICRA 2025! One paper (PSEC) is accepted by ICLR 2025!
• [2025.01] Guest Speaker sharing RSP at AI TIME AAAI Forum. [video(Chinese)]
• [2024.12] One paper (RSP) is accepted by AAAI 2025 as Oral Presentation!
• [2024.05] One paper (DecisionNCE) is accepted by ICML 2024!
• [2024.04] One Survey paper on Cross-Domain Embodied Policy Transfer is accepted by IJCAI 2024!
• [2024.01] One paper (CLIC) is accepted by ICRA 2024!
• [2023.10] Two papers (CLIC, SDM) are accepted by NeurIPS 2023 Symposium on ML4AD and ALOE Workshop!
• [2023.03] One paper ((Re)$^2$H2O) is accepted by IEEE IV 2023!
• [2023.02] Guest Speaker at 2023 AI TIME Youth PhD Talk Forum.
• [2022.11] Invited talk about our NeurIPS paper (H2O) at RLChina Academic Seminar! [video]
• [2022.11] Invited talk about our NeurIPS paper (H2O) at DI Lab’s group meeting!
• [2022.10] One paper is accepted by NeurIPS 2022 RL4RealLife Workshop and ML4AD Workshop!
• [2022.09] One paper (H2O) is accepted by NeurIPS 2022 as Spotlight!
• [2022.09] One paper (DMIL) is accepted by CoRL 2022!

📝 Publications

Preprints and Codebases

• Dongxiu Liu$^*$, Haoyi Niu$^*$✉, Peng Cheng, Yuan Gao, Xirui Kang, Sangli Teng, Koushil Sreenath, Xianyuan Zhan✉, ODEWorld: A Continuous Predictive Architecture via Physical-Time Flow, 2026. [Website]

• Yilin Wang$^*$, Shangzhe Li$^*$, Haoyi Niu, Zhiao Huang and Hao Su, A Recipe for Efficient Sim-to-Real Transfer in Manipulation with Online Imitation-Pretrained World Models, 2025.

• Xiangyu Zhu$^*$, Jianxiong Li$^*$, Haoyi Niu$^*$, Yinan Zheng$^*$, Peng Cheng$^*$, Wenjia Zhang$^*$, Haoran Xu$^*$ and Xianyuan Zhan, D2C: A Data-Driven Control Library Based on Reinforcement Learning, Paper coming soon, 2023. [Code][Documentation]

Conference Proceedings and Workshop Papers

• Haoyi Niu$^*$, Qimao Chen$^*$, Tenglong Liu, Jianxiong Li, Guyue Zhou, Yi Zhang, Jianming Hu and Xianyuan Zhan, xTED: Cross-Domain Adaptation via Diffusion-Based Trajectory Editing, International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS 2026). [Code][Website]
• Zhihao Wang$^*$, Jianxiong Li$^*$, Jinliang Zheng$^*$, Wencong Zhang, Dongxiu Liu, Yinan Zheng, Haoyi Niu, Junzhi Yu and Xianyuan Zhan, PhysiAgent: An Embodied Agent Framework in Physical World, the New in ML Workshop at ICML 2025, 2025.
• Dongxiu Liu$^*$, Haoyi Niu$^*$, Zhihao Wang, Jinliang Zheng, Yinan Zheng, Zhonghong Ou, Jianming Hu, Jianxiong Li, Xianyuan Zhan, Efficient Robotic Policy Learning via Latent Space Backward Planning, International Conference on Machine Learning (ICML 2025). [Code][Website]
• Haoyi Niu$^*$, Tianying Ji$^*$, Bingqi Liu, Haocheng Zhao, Xiangyu Zhu, Jianying Zheng, Pengfei Huang, Guyue Zhou, Jianming Hu and Xianyuan Zhan, H2O+: An Improved Framework for Hybrid Offline-and-Online RL with Dynamics Gaps, IEEE International Conference on Robotics and Automation (ICRA 2025). [Code][Wbsite]
• Tenglong Liu$^*$, Jianxiong Li$^*$, Yinan Zheng, Haoyi Niu, Yixing Lan, Xin Xu and Xianyuan Zhan, Skill Expansion and Composition in Parameter Space, International Conference on Learning Representations (ICLR 2025).
• Guan Wang$^*$, Haoyi Niu$^*$, Jianxiong Li, Li Jiang, Jianming Hu and Xianyuan Zhan, Are Expressive Models Truly Necessary for Offline RL?, Annual AAAI Conference on Artificial Intelligence (AAAI 2025), Oral Presentation. [Code][Twitter]
• Jianxiong Li, Jinliang Zheng, Yinan Zheng, Liyuan Mao, Xiao Hu, Sijie Cheng, Haoyi Niu, Jihao Liu, Yu Liu, Jingjing Liu, Ya-Qin Zhang and Xianyuan Zhan, DecisionNCE: Embodied Multimodal Representations via Implicit Preference Learning, International Conference on Machine Learning (ICML 2024). [Website][Code]
• Haoyi Niu, Jianming Hu, Guyue Zhou and Xianyuan Zhan, A Comprehensive Survey of Cross-Domain Policy Transfer for Embodied Agents, International Joint Conference on Artificial Intelligence (IJCAI 2024), Oral Presentation. [Repo]
• Haoyi Niu$^*$, Yizhou Xu$^*$, Xingjian Jiang and Jianming Hu, Continual Driving Policy Optimization with Closed-Loop Individualized Curricula, IEEE International Conference on Robotics and Automation (ICRA 2024). [Code]
• Haoyi Niu$^*$, Qimao Chen$^*$, Yingyue Li and Jianming Hu, Stackelberg Driver Model for Continual Policy Improvement in Scenario-Based Closed-Loop Autonomous Driving, NeurIPS 2023 Symposium on Machine Learning for Autonomous Driving & Agent Learning in Open-Endedness Workshop, 2023. [Code]
• Haoyi Niu$^*$, Kun Ren$^*$, Yizhou Xu, Ziyuan Yang, Yichen Lin, Yi Zhang and Jianming Hu, (Re)$^2$H2O: Autonomous Driving Scenario Generation via Reversely Regularized Hybrid Offline-and-Online Reinforcement Learning, IEEE Intelligent Vehicles Symposium 2023 (IV 2023). [Code]
• Haoyi Niu, Shubham Sharma, Yiwen Qiu, Ming Li, Guyue Zhou, Jianming Hu and Xianyuan Zhan, When to Trust Your Simulator: Dynamics-Aware Hybrid Offline-and-Online Reinforcement Learning, Thirty-sixth Conference on Neural Information Processing Systems (NeurIPS 2022), Spotlight. [Video][Code][Spotlight]
• Wenjia Zhang, Haoran Xu, Haoyi Niu, Peng Cheng, Ming Li, Heming Zhang, Guyue Zhou and Xianyuan Zhan, Discriminator-Guided Model-Based Offline Imitation Learning, 6th Conference on Robot Learning (CoRL 2022).
• Guan Wang$^*$, Haoyi Niu$^*$, Desheng Zhu, Jianming Hu, Xianyuan Zhan and Guyue Zhou, A Versatile and Efficient Reinforcement Learning Approach for Autonomous Driving, NeurIPS Machine Learning for Autonomous Driving Workshop, NeurIPS RL for RealLife Workshop, 2022. [Video][OneRL Code][NoGap Benchmark]
• Haoyi Niu, Jianming Hu, Zheyu Cui and Yi Zhang, Tactical Decision Making for Emergency Vehicles Based on A Combinational Learning Method, the 20th and 21st Joint COTA International Conference of Transportation Professionals (CICTP 2020-21), Best Paper Award.
• Haoyi Niu, Jianming Hu, Zheyu Cui and Yi Zhang, DR$^2$L: Surfacing Corner Cases to Robustify Autonomous Driving via Domain Randomization Reinforcement Learning, the 5th International Conference on Computer Science and Application Engineering (CSAE 2021), Oral Presentation.
• Qingyu Song, Ruibo Ming, Jianming Hu, Haoyi Niu and Mingyang Gao, Graph Attention Convolutional Network: Spatiotemporal Modeling for Urban Traffic Prediction, IEEE 23rd International Conference on Intelligent Transportation Systems (ITSC 2020).

🎖 Honors and Awards

• [2026.03] Chinese Association of Automation (CAA) Outstanding Master Dissertation
• [2025.06] Beijing Outstanding Graduate Award
• [2025.06] Tsinghua Outstanding Master Dissertation (top 5%)
• [2025.06] Tsinghua Outstanding Master Graduate Award (top 1%)
• [2025.06] Outstanding Master Graduate Award of Department of Automation, Tsinghua University
• [2025.01] AAAI 2025 Oral Presentation Award
• [2024.11] Tsinghua First-Class Comprehensive Excellence Award for Graduate Students (Fang Chongzhi Development Fund Scholarship)
• [2024.01] Zheng Weimin Scholarship
• [2023.10] Tsinghua First-Class Comprehensive Excellence Award for Graduate Students (Lingjun Investment Scholarship)
• [2022.12] NeurIPS 2022 Spotlight Award
• [2022.06] Tsinghua Outstanding Bachelor Dissertation (top 5%)
• [2022.06] Tsinghua Excellent Bachelor Graduate Award (top 10%)
• [2022.06] Outstanding Bachelor Graduate Award of Department of Automation, Tsinghua University
• [2021.12] Best Paper Award at COTA International Conference of Transportation Professionals (CICTP)
• [2021.10] Tsinghua Scientific and Technological Innovation Excellence Award
• [2021.09] Tsinghua “Future Scholar” Scientific Research Grant (9/3850 in Tsinghua Class of 2022)
• [2021.04] Second Prize in Tsinghua “Challenge Cup” Academic Science and Technology Competition
• [2020.12] First Prize in Tsinghua Excellent Student Research Training (SRT) Program (top 1%, 12/1250)
• [2020.10] Tsinghua Comprehensive Excellence Award (top 10%, 15/179)
• [2020.10] Fang Chongzhi Scholarship (highest scholarship for juniors in Dept. of Automation, 1/179)
• [2019.12] First Prize in Group A (Non-Physics Major) of the National Collegiate Physics Competition

📖 Educations

• 2025.08 - Present, PhD, UC Berkeley, CA, USA.
• 2022.08 - 2025.06, Master, Department of Automation, Tsinghua University, Beijing, China.
• 2018.08 - 2022.06, Undergraduate, Department of Automation, Tsinghua Univeristy, Beijing, China.

💻 Internships

• 2021.03 - 2025.06, AIR-DREAM Lab of Institute for AI Industry Research (AIR), Tsinghua University, Beijing, China. (Advisor: Prof.Xianyuan Zhan)

🧑‍🎨 Services

Reviewer for Conferences (ITSC, ICRA, IROS, IJCAI, NeurIPS, ICML) and Journals (IEEE Robotics and Automation Letters, Transactions on Machine Learning Research, IEEE Transactions on Intelligent Vehicles, Transportation Research Part C).