Tonghe Zhang

News

Talks

Research

Preprints

• BFM-Zero: A Promptable Behavioral Foundation Model for Humanoid Control Using Unsupervised Reinforcement Learning

  Yitang Li, Zhengyi Luo, Tonghe Zhang, Cunxi Dai, Anssi Kanervisto, Andrea Tirinzoni, Haoyang Weng, Kris Kitani, Mateusz Guzek, Ahmed Touati, Alessandro Lazaric, Matteo Pirotta, Guanya Shi.

  Preprint.

  TL;DR: Building behavior foundation model for Humanoid loco-manipulation via unsupervised sim2real RL and sampling.

• Sample-Efficient Reinforcement Learning of Flow Policies via Sequence Modeling

  Yixian Zhang, Shu'ang Yu, Tonghe Zhang, Mo Guang, Haojia Hui, Kaiwen Long, Yu Wang, Chao Yu, Wenbo Ding.

  Preprint.

  TL;DR: Training flow matching policies via off-policy RL by treating the process as RNN/Transformer.

Publications

• 
  ReinFlow: Fine-tuning Flow Matching Policy with Online Reinforcement Learning

  Tonghe Zhang, Chao Yu, Sichang Su, Yu Wang.

  The Thirty-Ninth Annual Conference on Neural Information Processing Systems (NeurIPS 2025).

  TL;DR: Injecting learnable noise to fine-tune flow matching policies (e.g., 1-Rectified Flow and Shortcut Models) with policy gradient.

  NeurIPS Homepage Website arXiv Google Scholar GitHub WandB HuggingFace BibTex

  @misc{zhang2025reinflowfinetuningflowmatching,
        title={ReinFlow: Fine-tuning Flow Matching Policy with Online Reinforcement Learning}, 
        author={Tonghe Zhang and Chao Yu and Sichang Su and Yu Wang},
        year={2025},
        eprint={2505.22094},
        archivePrefix={arXiv},
        primaryClass={cs.RO},
        url={https://arxiv.org/abs/2505.22094}, 
  }