GR-Dexter Technical Report
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911361935605760 |
|---|---|
| author | Wen, Ruoshi Chen, Guangzeng Cui, Zhongren Du, Min Gou, Yang Han, Zhigang Huang, Liqun Lei, Mingyu Li, Yunfei Li, Zhuohang Liu, Wenlei Liu, Yuxiao Ma, Xiao Niu, Hao Ouyang, Yutao Ren, Zeyu Shi, Haixin Xu, Wei Zhang, Haoxiang Zhang, Jiajun Zhang, Xiao Zheng, Liwei Zhong, Weiheng Zhou, Yifei Zhu, Zhengming Li, Hang |
| author_facet | Wen, Ruoshi Chen, Guangzeng Cui, Zhongren Du, Min Gou, Yang Han, Zhigang Huang, Liqun Lei, Mingyu Li, Yunfei Li, Zhuohang Liu, Wenlei Liu, Yuxiao Ma, Xiao Niu, Hao Ouyang, Yutao Ren, Zeyu Shi, Haixin Xu, Wei Zhang, Haoxiang Zhang, Jiajun Zhang, Xiao Zheng, Liwei Zhong, Weiheng Zhou, Yifei Zhu, Zhengming Li, Hang |
| contents | Vision-language-action (VLA) models have enabled language-conditioned, long-horizon robot manipulation, but most existing systems are limited to grippers. Scaling VLA policies to bimanual robots with high degree-of-freedom (DoF) dexterous hands remains challenging due to the expanded action space, frequent hand-object occlusions, and the cost of collecting real-robot data. We present GR-Dexter, a holistic hardware-model-data framework for VLA-based generalist manipulation on a bimanual dexterous-hand robot. Our approach combines the design of a compact 21-DoF robotic hand, an intuitive bimanual teleoperation system for real-robot data collection, and a training recipe that leverages teleoperated robot trajectories together with large-scale vision-language and carefully curated cross-embodiment datasets. Across real-world evaluations spanning long-horizon everyday manipulation and generalizable pick-and-place, GR-Dexter achieves strong in-domain performance and improved robustness to unseen objects and unseen instructions. We hope GR-Dexter serves as a practical step toward generalist dexterous-hand robotic manipulation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_24210 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | GR-Dexter Technical Report Wen, Ruoshi Chen, Guangzeng Cui, Zhongren Du, Min Gou, Yang Han, Zhigang Huang, Liqun Lei, Mingyu Li, Yunfei Li, Zhuohang Liu, Wenlei Liu, Yuxiao Ma, Xiao Niu, Hao Ouyang, Yutao Ren, Zeyu Shi, Haixin Xu, Wei Zhang, Haoxiang Zhang, Jiajun Zhang, Xiao Zheng, Liwei Zhong, Weiheng Zhou, Yifei Zhu, Zhengming Li, Hang Robotics Vision-language-action (VLA) models have enabled language-conditioned, long-horizon robot manipulation, but most existing systems are limited to grippers. Scaling VLA policies to bimanual robots with high degree-of-freedom (DoF) dexterous hands remains challenging due to the expanded action space, frequent hand-object occlusions, and the cost of collecting real-robot data. We present GR-Dexter, a holistic hardware-model-data framework for VLA-based generalist manipulation on a bimanual dexterous-hand robot. Our approach combines the design of a compact 21-DoF robotic hand, an intuitive bimanual teleoperation system for real-robot data collection, and a training recipe that leverages teleoperated robot trajectories together with large-scale vision-language and carefully curated cross-embodiment datasets. Across real-world evaluations spanning long-horizon everyday manipulation and generalizable pick-and-place, GR-Dexter achieves strong in-domain performance and improved robustness to unseen objects and unseen instructions. We hope GR-Dexter serves as a practical step toward generalist dexterous-hand robotic manipulation. |
| title | GR-Dexter Technical Report |
| topic | Robotics |
| url | https://arxiv.org/abs/2512.24210 |