avatar

Weirui Ye (叶葳蕤)

Postdoc
CSAIL, MIT
yeweirui16 (at) gmail (dot) com


About Me

I am a Postdoc Fellow in LIS Group at MIT CSAIL, working with Prof. Leslie Pack Kaelbling, Tomás Lozano-Pérez at MIT and Yilun Du at Harvard. Previously, I received my Ph.D. from Tsinghua University under the supervision of Prof. Yang Gao, and my B.Sc. from Tsinghua University. During my Ph.D., I also worked as a visiting scholar at UC Berkeley, advised by Prof. Pieter Abbeel.

My research goal is to develop human-level policy learning algorithms for robots in learning efficiency and generalization. I am interested in Reinforcement Learning, Generative World Models and Lifelong Learning for Robots.

News from 2024

Selected Publications

* equal contribution · † co-advisor
  1. ArXiv
    DriftWorld: Fast World Modeling through Drifting
    Susie Lu, Haonan Chen, Weirui Ye, Yilun Du
    ArXiv, 2026.
    PDF Code Website

  2. ArXiv
    Temporal Backtracking Search for Test-time Generative Video Reasoning
    Sejoon Jun, Zheng Ding, Chloe H. Su, Weirui Ye†, Yilun Du†
    ArXiv, 2026.
    PDF Website

  3. ArXiv
    Scaling Tasks, Not Samples: Mastering Humanoid Control through Multi-Task Model-Based Reinforcement Learning
    Shaohuai Liu*, Weirui Ye*, Yilun Du†, Le Xie†
    ArXiv, 2026.
    PDF Code Website

  4. ICLR
    TreeGRPO: Tree-Advantage GRPO for Online RL Post-Training of Diffusion Models
    Zheng Ding*, Weirui Ye*
    International Conference on Learning Representations (ICLR), 2026.
    PDF Code Website

  5. ArXiv
    Video2Policy: Scaling up Manipulation Tasks in Simulation through Internet Videos
    Weirui Ye, Fangchen Liu, Zheng Ding, Yang Gao, Oleh Rybkin, Pieter Abbeel
    ArXiv, 2025.
    PDF Website

  6. CoRL
    Reinforcement Learning with Foundation Priors: Let the Embodied Agent Efficiently Learn on Its Own
    Weirui Ye, Yunsheng Zhang, Haoyang Weng, Xianfan Gu, Shengjie Wang, Tong Zhang, Mengchen Wang, Pieter Abbeel, Yang Gao
    Conference on Robot Learning (CoRL), 2024.
    PDF Code Website   Oral, 4%

  7. ICML
    EfficientZero V2: Mastering Discrete and Continuous Control with Limited Data
    Shengjie Wang*, Shaohuai Liu*, Weirui Ye*, Jiacheng You, Yang Gao
    International Conference on Machine Learning (ICML), 2024.
    PDF Code   Spotlight, 3.5%

  8. ICLR
    Become a Proficient Player through Watching Pure Videos
    Weirui Ye*, Yunsheng Zhang*, Pieter Abbeel, Yang Gao
    International Conference on Learning Representations (ICLR), 2023.
    PDF Code

  9. NeurIPS
    Spending Thinking Time Wisely: Accelerating MCTS with Virtual Expansions
    Weirui Ye, Pieter Abbeel, Yang Gao
    Conference on Neural Information Processing Systems(NeurIPS), 2022.
    PDF

  10. NeurIPS
    Mastering Atari Games with Limited Data
    Weirui Ye, Shaohuai Liu, Thanard Kurutach, Pieter Abbeel, Yang Gao
    Conference on Neural Information Processing Systems(NeurIPS), 2021.
    PDF Code Towards Data Science Interview Tsinghua News (Chinese) YouTube


Professional Services

Invited Talks

Workshop Organization

Reviewers | PC Member

Teaching Assistant

Student Counselor


Selected Awards


Powered by Jekyll and Minimal Light theme.