π¨π»βπ» About Me
Hi! I am Dongxiu Liu, a co-founder of Basal Intelligence. Before that, I was a long-term research intern at the Institute for AI Industry Research (AIR), Tsinghua University, where I was fortunate to work closely with Prof. Xianyuan Zhan. I received my bachelorβs and masterβs degrees from Beijing University of Posts and Telecommunications.
I focus on robot learning, with the long-term goal of building embodied agents that can understand and interact with the physical world, while generalizing across tasks and embodiments.
π₯ News
- 2026.09: Β ππ ODEWorld has been accepted by NeurIPS 2026!
- 2026.09: Β ππ X-Tokenizer has been accepted by CoRL 2026!
- 2026.05: Β ππ An Empirical Study on action space design has been accepted by ICML 2026 and won the π Best Paper Award at the ES-Reasoning Workshop @ ICLR 2026!
- 2026.01: Β ππ X-VLA has been accepted by ICLR 2026!
- 2025.12: Β ππ UAGLNet has been accepted by IEEE TGRS!
- 2025.10: Β ππ X-VLA won the π Championship at AgiBot World Challenge @ IROS 2025!
- 2025.05: Β ππ LBP has been accepted by ICML 2025!
- 2025.02: Β ππ UniAct has been accepted by CVPR 2025!
π Publications
* denotes equal contribution.
- 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. In Conference on Neural Information Processing Systems (NeurIPS 2026). [Code][Website]
- Miracle Kang$^*$, Lights Shi$^*$, Lucy Liang, Roy Gan, Dongxiu Liu, Pushi Zhang, Sylas Chen, Shawn Qin, Yinan Zheng, Jinliang Zheng, Hao Wang, Xianyuan Zhan, Hang Su. X-Tokenizer: A Multimodal Action Tokenizer for Vision-Language-Action Model Pretraining. In Conference on Robot Learning (CoRL 2026). [Code][Website]
- Yuchun Feng$^*$, Jinliang Zheng$^*$, Zhihao Wang, Dongxiu Liu, Jianxiong Li, Jiangmiao Pang, Tai Wang, Xianyuan Zhan. Demystifying Action Space Design for Robotic Manipulation Policies. In International Conference on Machine Learning (ICML 2026).
- Jinliang Zheng$^*$, Jianxiong Li$^*$, Zhihao Wang, Dongxiu Liu, Xirui Kang, Yuchun Feng, Yinan Zheng, Jiayin Zou, Yilun Chen, Jia Zeng, Ya-Qin Zhang, Jiangmiao Pang, Jingjing Liu, Tai Wang, Xianyuan Zhan. X-VLA: Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Model. In International Conference on Learning Representations (ICLR 2026). [Code][Website]
- Siyuan Yao, Dongxiu Liu, Taotao Li, Shengjie Li, Wenqi Ren, Xiaochun Cao. UAGLNet: Uncertainty-Aggregated Global-Local Fusion Network with Cooperative CNN-Transformer for Building Extraction. In IEEE Transactions on Geoscience and Remote Sensing (IEEE TGRS). [Code]
- Zhihao Wang$^*$, Jianxiong Li$^*$, Jinliang Zheng$^*$, Wencong Zhang, Dongxiu Liu, Yinan Zheng, Haoyi Niu, Junzhi Yu, Xianyuan Zhan. PhysiAgent: An Embodied Agent Framework in Physical World. In New in ML Workshop (ICML 2025 Workshop).
- 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. In International Conference on Machine Learning (ICML 2025). [Code][Website]
- Jinliang Zheng$^*$, Jianxiong Li$^*$, Dongxiu Liu$^*$, Yinan Zheng, Zhihao Wang, Zhonghong Ou, Yu Liu, Jingjing Liu, Ya-Qin Zhang, Xianyuan Zhan. Universal Actions for Enhanced Embodied Foundation Models. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2025). [Code][Website]
π Honors and Awards
- 2025.10 π Champion at AGIBOT World Challenge (Manipulation Track) @ IROS 2025.
- 2023.06 Outstanding Graduate of Beijing.
- 2021.09 Overall Top 0.1% at CCF Certified Software Professional (CSP).
- 2021.05 π Gold Medal at International Collegiate Programming Contest Asia Regional Contest (ICPC).
- 2021.04 Finalist at International Mathematical Contest in Modeling (MCM).