I am a final-year Ph.D. candidate in the School of Computer Science at PKU, advised by Prof. Hao Dong. I received my bachelor’s degree in 2021 from the Turing Class at PKU.
In 2025, I worked with Jianlan Luo at AgiBot on confidential, cutting-edge research projects. My research focuses on real-world reinforcement learning and tactile robotic systems, with a vision to build self-evolving robotic foundation models through autonomous data-flywheels.
@article{wu2025simlauncher,video={https://www.dropbox.com/scl/fi/5yif7dl2bz4pcv2uw9s03/Training-Timelapse.mp4?rlkey=iga15684bqfy2fz9fwk6edsay&st=mryoerqt&dl=0},title={SimLauncher: Launching Sample-Efficient Real-world Robotic Reinforcement Learning via Simulation Pre-training},author={Wu*, Mingdong and Wu*, Lehong and Wu*, Yizhuo and Huang, Weiyao and Fan, Hongwei and Hu, Zheyuan and Geng, Haoran and Li, Jinzhou and Ying, Jiahe and Yang, Long and others},journal={IEEE/RSJ International Conference on Intelligent Robots and Systems},year={2025}}
CoRL 2025
UniTac2Pose: A Unified Approach Learned in Simulation for Generalizable Visuotactile In-hand Pose Estimation
Mingdong Wu*, Long Yan*, Jin Liu*, Weiyao Huang, Lehong Wu, Zelin Chen, Daolin Ma, and Hao Dong
@article{wu2025unitac2pose,title={UniTac2Pose: A Unified Approach Learned in Simulation for Generalizable Visuotactile In-hand Pose Estimation},author={Wu*, Mingdong and Yan*, Long and Liu*, Jin and Huang, Weiyao and Wu, Lehong and Chen, Zelin and Ma, Daolin and Dong, Hao},journal={Conference on Robot Learning},year={2025}}
IROS 2025
Adaptive Visuo-Tactile Fusion with Predictive Force Attention for Dexterous Manipulation
@article{li2025adaptive,title={Adaptive Visuo-Tactile Fusion with Predictive Force Attention for Dexterous Manipulation},author={Li*, Jinzhou and Wu*, Tianhao and Zhang, Jiyao and Chen, Zeyuan and Jin, Haotian and Wu, Mingdong and Shen, Yujun and Yang, Yaodong and Dong, Hao},journal={IEEE/RSJ International Conference on Intelligent Robots and Systems},year={2025}}
ICRA 2025
Canonical representation and force-based pretraining of 3d tactile for dexterous visuo-tactile policy learning
@article{wu2024canonical,title={Canonical representation and force-based pretraining of 3d tactile for dexterous visuo-tactile policy learning},author={Wu, Tianhao and Li*, Jinzhou and Zhang*, Jiyao and Wu, Mingdong and Dong, Hao},journal={IEEE International Conference on Robotics and Automation},year={2025}}
2024
ECCV 2024
Omni6DPose: A Benchmark and Model for Universal 6D Object Pose Estimation and Tracking
Jiyao Zhang*, Weiyao Huang*, Bo Peng*, Mingdong Wu, Fei Hu, Zijian Chen, Bo Zhao, and Hao Dong
@article{zhang2024omni6dpose,title={Omni6DPose: A Benchmark and Model for Universal 6D Object Pose Estimation and Tracking},author={Zhang*, Jiyao and Huang*, Weiyao and Peng*, Bo and Wu, Mingdong and Hu, Fei and Chen, Zijian and Zhao, Bo and Dong, Hao},journal={European Conference on Computer Vision},year={2024}}
RAL 2024
Distilling Functional Rearrangement Priors from Large Models
Yiming Zeng*, Mingdong Wu*, Long Yang, Jiyao Zhang, Hao Ding, Hui Cheng, and Hao Dong
@article{zeng2023distilling,title={Distilling Functional Rearrangement Priors from Large Models},author={Zeng*, Yiming and Wu*, Mingdong and Yang, Long and Zhang, Jiyao and Ding, Hao and Cheng, Hui and Dong, Hao},journal={IEEE Robotics and Automation Letters},year={2024}}
2023
NeurIPS 2023
GenPose: Generative Category-level Object Pose Estimation via Diffusion Models
We explore a pure generative approach to tackle the multi-hypothesis issue in 6D Category-level Object Pose Estimation. The key idea is to generate pose candidates using a score-based diffusion model and filter out outliers using an energy-based diffusion model. By aggregating the remaining candidates, we can obtain a robust and high-quality output pose.
@article{zhang2023genpose,news={机器之心},news_link={https://mp.weixin.qq.com/s/RYV_aap9eYtwX_4_Ghr5Vw},sota_link={https://paperswithcode.com/sota/6d-pose-estimation-using-rgbd-on-real275?p=genpose-generative-category-level-object-pose},sota_badge={https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/genpose-generative-category-level-object-pose/6d-pose-estimation-using-rgbd-on-real275},star={https://img.shields.io/github/stars/Jiyao06/GenPose?style=social&label=Code+Stars},title={GenPose: Generative Category-level Object Pose Estimation via Diffusion Models},author={Zhang*, Jiyao and Wu*, Mingdong and Dong, Hao},journal={Thirty-seventh Conference on Neural Information Processing Systems},year={2023}}
NeurIPS 2023
Learning Score-based Grasping Primitive for Human-assisting Dexterous Grasping
@article{wu2023learning,news={新智元},news_link={https://mp.weixin.qq.com/s/hpzZWMizR8tPSGIvGVjPoA},star={https://img.shields.io/github/stars/tianhaowuhz/human-assisting-dex-grasp?style=social&label=Code+Stars},title={Learning Score-based Grasping Primitive for Human-assisting Dexterous Grasping},author={Wu*, Tianhao and Wu*, Mingdong and Zhang, Jiyao and Gan, Yunchong and Dong, Hao},journal={Thirty-seventh Conference on Neural Information Processing Systems},year={2023}}
2022
NeurIPS 2022
TarGF: Learning Target Gradient Field to Rearrange Objects without Explicit Goal Specification
We study object rearrangement without explicit goal specification. The agent is given examples from a target distribution and aims at rearranging objects to increase the likelihood of the distribution. Our key idea is to learn a target gradient field that indicates the fastest direction to increase the likelihood from examples via score-matching.
@inproceedings{wu2022targf,title={Tar{GF}: Learning Target Gradient Field to Rearrange Objects without Explicit Goal Specification},author={Wu*, Mingdong and Zhong*, Fangwei and Xia, Yulong and Dong, Hao},booktitle={Advances in Neural Information Processing Systems},editor={Oh, Alice H. and Agarwal, Alekh and Belgrave, Danielle and Cho, Kyunghyun},year={2022},url={https://openreview.net/forum?id=Euv1nXN98P3}}
