Yitao Liu

In me the tiger sniffs the rose
xinjiang_square.jpeg

I am a first-year PhD student at the University of Hong Kong, advised by Tao Yu in XLANG Lab. Previously, I worked with Karthik Narasimhan in Princeton NLP Group focusing on language agents in 2024. I was fortuntely mentored by Shunyu Yao during my visiting to Princeton University. The goal of my research is to build an intelligent language agent that can interact with various environments effectively and efficiently, including digital ones (web, computer, etc.) and physical ones (real-world robot control, etc.). Feel free to reach out to me if you are interested in my work and have any questions or ideas to discuss with!

news

May 26, 2025 CER was accepted by ACL 2025!
Apr 9, 2025 Thanks Shunyu Yao for presenting an inspiring blog The Second Half, which contains a lot of insights gained from discussion and brainstorming during our work CER. Also great thanks for his mentorship during this project!
Jul 9, 2024 OpenAgents was accepted by COLM 2024
Jan 16, 2024 Two papers accepted by ICLR 2024 as Spotlight
Apr 16, 2023 The webpage is born

selected publications

* equal contribution
  1. roboinf.png
    RoboInF: Scaling Robot Manipulation Data in Simulation for Embodied Instruction Following
    NLP RL
    XLANG Lab
    XLANG Lab Blog, May 2026
    Contribution: Project Lead; Full paper and code are on the way.
  2. qwenvla.png
    Qwen-VLA: Unifying Vision-Language-Action Modeling across Tasks, Environments, and Robot Embodiments
    NLP RL
    Qiuyue Wang, Mingsheng Li, Jian Guan, Jinhui Ye, Sicheng Xie, Yitao Liu, Junhao Chen, Zhixuan Liang, Jie Zhang, Xintong Hu, Xuhong Huang, Pei Lin, Junyang Lin, Dayiheng Liu, Shuai Bai, Jingren Zhou, Jiazhao Zhang, Haoqi Yuan, Gengze Zhou, Hang Yin, Ye Wang, Yiyang Huang, Zixing Lei, Wujian Peng, Delin Chen, Yingming Zheng, Jingyang Fan, Xianwei Zhuang, Xin Zhou, Haoyang Li, Anzhe Chen, Tong Zhang, Xuejing Liu, Yuchong Sun, Ruizhe Chen, Zhaohai Li, Chenxu Lü, Zhibo Yang, Tao Yu, and Xionghui Chen
    Preprint, May 2026
    Contribution: Core Contributor; Lead the simulation data scaling pipeline
  3. finevla.png
    FineVLA: Fine-Grained Instruction Alignment for Steerable Vision-Language-Action Policies
    NLP RL
    Xintong Hu*, Xuhong Huang*, Jinyu Zhang, Yutong Yao, Yuchong Sun, Qiuyue Wang, Mingsheng Li, Sicheng Xie, Yitao Liu, Junhao Chen, Yixuan Chen, Yingming Zheng, Shuai Bai, and Tao Yu
    Preprint, May 2026
  4. CER_overview.png
    Contextual Experience Replay for Self-Improvement of Language Agents
    NLP
    Yitao Liu, Chenglei Si, Karthik Narasimhan, and Shunyu Yao
    ACL, May 2025
  5. openagents_overview.png
    OpenAgents: An Open Platform for Language Agents in the Wild
    NLP
    Tianbao Xie*, Fan Zhou*, Zhoujun Cheng*, Peng Shi*, Luoxuan Weng*, Yitao Liu*, Toh Jing Hua, Junning Zhao, Qian Liu, Che Liu, Leo Z. Liu, Yiheng Xu, Hongjin Su, Dongchan Shin, Caiming Xiong, and Tao Yu
    Conference on Language Modeling (COLM), May 2024
    Contribution: Co-lead the web agent and plug-in agent framework
  6. osworld_overview.png
    OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments
    NLP
    Tianbao Xie, Danyang Zhang, Jixuan Chen, Xiaochuan Li, Siheng Zhao, Ruisheng Cao, Toh Jing Hua, Zhoujun Cheng, Dongchan Shin, Fangyu Lei, Yitao Liu, Yiheng Xu, Shuyan Zhou, Silvio Savarese, Caiming Xiong, Victor Zhong, and Tao Yu
    Thirty-eighth Conference on Neural Information Processing Systems (NeurIPS), May 2024
  7. charxiv.png
    CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs
    NLP
    Zirui Wang, Mengzhou Xia, Luxi He, Howard Chen, Yitao Liu, Richard Zhu, Kaiqu Liang, Xindi Wu, Haotian Liu, Sadhika Malladi, Alexis Chevalier, Sanjeev Arora, and Danqi Chen
    May 2024
    Contribution: I annotated and verified nearly half of the benchmark questions. I also co-developed the question-quality criteria and evaluation metrics, contributed to paper writing, and curated representative examples for the showcase.
  8. text2reward.png
    Text2Reward: Automated Dense Reward Function Generation for Reinforcement Learning
    NLP RL
    Tianbao Xie*Siheng Zhao*Chen Henry WuYitao Liu, Qian Luo, Victor Zhong, Yanchao Yang, and Tao Yu
    International Conference on Learning Representations (ICLR), May 2024
    Spotlight
  9. lemur.png
    Lemur: Harmonizing Natural Language and Code for Language Agents
    NLP
    Yiheng Xu*, Hongjin Su*, Chen Xing*, Boyu Mi, Qian Liu, Weijia Shi, Binyuan Hui, Fan Zhou, Yitao LiuTianbao Xie, Zhoujun Cheng, Siheng Zhao, Lingpeng Kong, Bailing Wang, Caiming Xiong, and Tao Yu
    International Conference on Learning Representations (ICLR), May 2024
    Contribution: Data filtering and evaluation of WebArena
    Spotlight

friends

alphabetical order
Liheng Chen
Meitong Liu
Tianbao Xie
Yaowen Ye