Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2601.18692 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910033400299520 |
|---|---|
| author | Wu, Wei Lu, Fan Wang, Yunnan Yang, Shuai Liu, Shi Wang, Fangjing Zhu, Qian Sun, He Wang, Yong Ma, Shuailei Ren, Yiyu Zhang, Kejia Yu, Hui Zhao, Jingmei Zhou, Shuai Qiu, Zhenqi Xiong, Houlong Wang, Ziyu Wang, Zechen Cheng, Ran Li, Yong-Lu Huang, Yongtao Zhu, Xing Shen, Yujun Zheng, Kecheng |
| author_facet | Wu, Wei Lu, Fan Wang, Yunnan Yang, Shuai Liu, Shi Wang, Fangjing Zhu, Qian Sun, He Wang, Yong Ma, Shuailei Ren, Yiyu Zhang, Kejia Yu, Hui Zhao, Jingmei Zhou, Shuai Qiu, Zhenqi Xiong, Houlong Wang, Ziyu Wang, Zechen Cheng, Ran Li, Yong-Lu Huang, Yongtao Zhu, Xing Shen, Yujun Zheng, Kecheng |
| contents | Offering great potential in robotic manipulation, a capable Vision-Language-Action (VLA) foundation model is expected to faithfully generalize across tasks and platforms while ensuring cost efficiency (e.g., data and GPU hours required for adaptation). To this end, we develop LingBot-VLA with around 20,000 hours of real-world data from 9 popular dual-arm robot configurations. Through a systematic assessment on 3 robotic platforms, each completing 100 tasks with 130 post-training episodes per task, our model achieves clear superiority over competitors, showcasing its strong performance and broad generalizability. We have also built an efficient codebase, which delivers a throughput of 261 samples per second with an 8-GPU training setup, representing a 1.5~2.8$\times$ (depending on the relied VLM base model) speedup over existing VLA-oriented codebases. The above features ensure that our model is well-suited for real-world deployment. To advance the field of robot learning, we provide open access to the code, base model, and benchmark data, with a focus on enabling more challenging tasks and promoting sound evaluation standards. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_18692 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | A Pragmatic VLA Foundation Model Wu, Wei Lu, Fan Wang, Yunnan Yang, Shuai Liu, Shi Wang, Fangjing Zhu, Qian Sun, He Wang, Yong Ma, Shuailei Ren, Yiyu Zhang, Kejia Yu, Hui Zhao, Jingmei Zhou, Shuai Qiu, Zhenqi Xiong, Houlong Wang, Ziyu Wang, Zechen Cheng, Ran Li, Yong-Lu Huang, Yongtao Zhu, Xing Shen, Yujun Zheng, Kecheng Robotics Computer Vision and Pattern Recognition Offering great potential in robotic manipulation, a capable Vision-Language-Action (VLA) foundation model is expected to faithfully generalize across tasks and platforms while ensuring cost efficiency (e.g., data and GPU hours required for adaptation). To this end, we develop LingBot-VLA with around 20,000 hours of real-world data from 9 popular dual-arm robot configurations. Through a systematic assessment on 3 robotic platforms, each completing 100 tasks with 130 post-training episodes per task, our model achieves clear superiority over competitors, showcasing its strong performance and broad generalizability. We have also built an efficient codebase, which delivers a throughput of 261 samples per second with an 8-GPU training setup, representing a 1.5~2.8$\times$ (depending on the relied VLM base model) speedup over existing VLA-oriented codebases. The above features ensure that our model is well-suited for real-world deployment. To advance the field of robot learning, we provide open access to the code, base model, and benchmark data, with a focus on enabling more challenging tasks and promoting sound evaluation standards. |
| title | A Pragmatic VLA Foundation Model |
| topic | Robotics Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2601.18692 |