Built with and by Teycir Ben Soltane•
How to Use•FAQ•GitHub•arXiv.org•
Share:
ArXivCSExplorer
☆☆Bookmarks🏆RSSHow to UseFAQ
Home/Authors/Nan Xue

Nan Xue

3 indexed papers

Recent (6 mo)
3
With code
0
Influential cites
0
Benchmarked
0

Publications per year

3
26

Top categories

Vision×3Robotics×1

Frequent co-authors

Yujun Shen3×
Yinghao Xu3×
Jiahao Shao2×
Shuailei Ma2×
Jiaqi Liao2×
Zifan Shi2×

Research Timeline

2026
Scaling Mixture-of-Experts Video Pretraining for Embodied Intelligence

This paper introduces LingBot-Video, a video pretraining paradigm for embodied intelligence using a DiT-based approach, Mixture-of-Experts framework, and extensive robot-oriented data.

Native Video-Action Pretraining for Generalizable Robot Control

This paper introduces LingBot-VA 2.0, a video-action foundation model designed for embodiment, with semantic visual-action tokenization, causal pretraining, sparse MoE backbone, and enhanced asynchronous inference.

Persistent Planar Memory for Video World Models

This paper presents PPMem, a video world model that uses plane primitives as persistent 3D memory to generate and explore environments at scale.

Highlighted terms show continued research focus across papers

Papers

cs.CVEmpiricalRecentOct 16, 2026

Persistent Planar Memory for Video World Models

Yuze He, Bin Tan, Zelin Gao, Yujun Shen +3 more

This paper presents PPMem, a video world model that uses plane primitives as persistent 3D memory to generate and explore environments at scale.

View →
cs.ROcs.CVEmpiricalRecent
Jul 9, 2026

Native Video-Action Pretraining for Generalizable Robot Control

Qihang Zhang, Lin Li, Luyao Zhang, Shuai Yang +25 more

This paper introduces LingBot-VA 2.0, a video-action foundation model designed for embodiment, with semantic visual-action tokenization, causal pretraining, sparse MoE backbone, and enhanced asynchron…

View →
cs.CVEmpiricalRecentJul 8, 2026

Scaling Mixture-of-Experts Video Pretraining for Embodied Intelligence

Shuailei Ma, Jiaqi Liao, Xinyang Wang, Jingjing Wang +23 more

This paper introduces LingBot-Video, a video pretraining paradigm for embodied intelligence using a DiT-based approach, Mixture-of-Experts framework, and extensive robot-oriented data.

View →