Preprint

Learning Soccer Skills for Humanoid Robots: A Progressive Perception-Action Framework
arXiv preprint
Perception-Action integrated Decision-making (PAiD), a progressive framework for humanoid soccer skills.
Jipeng Kong , Xinzhe Liu , Yuhang Lin , Jinrui Han , Sören Schwertfeger , Chenjia Bai✉ , Xuelong Li
PDF Cite Project 公众号
Learning Soccer Skills for Humanoid Robots: A Progressive Perception-Action Framework
DeCoNav: Dialog enhanced Long-Horizon Collaborative Vision-Language Navigation
arXiv preprint
DeCoNav introduces event-triggered dialogue and dynamic replanning to improve long-horizon collaborative VLN in synchronized multi-robot settings.
Sunyao Zhou , Yunzi Wu , Tianhang Wang , Xinhai Li , Guang Chen , Lizheng Liu , Chenjia Bai , Xuelong Li
PDF Cite
DeCoNav: Dialog enhanced Long-Horizon Collaborative Vision-Language Navigation
SpaceVLN: A Zero-Shot Vision-and-Language Navigation Agent with Online Spatial Cognitive Memory and Reasoning
arXiv preprint
SpaceVLN introduces online Spatial Cognitive Memory and Task-Guided Spatial Reasoning for unified zero-shot VLN and Object-Goal Navigation.
Yucheng Deng , Pingrui Lai , Xinhai Li , Chenjia Bai , Xiaoheng Deng , Chengnuo Sun , Xuelong Li✉ , Hua Yang✉
PDF Cite Project
SpaceVLN: A Zero-Shot Vision-and-Language Navigation Agent with Online Spatial Cognitive Memory and Reasoning
UMI-Bench 1.0: An Open and Reproducible Real-World Benchmark for Tabletop Robotic Manipulation with UMI Data
arXiv preprint
UMI-Bench 1.0 provides an open, reproducible real-world benchmark for standardized evaluation of UMI-style tabletop manipulation policies.
Shi Jin , Yuntian Wang , Yuhui Duan , Di Wu , Gaoqi Dong , Xiaohang Liu , Xiaotong Li , Hongfei Jia , Zehao Zhang , Tianyu Wang , Zhongjie Jia , Yuanqi Yao , Chenjia Bai , Zhaxizhuoma✉ , Siao Liu✉ , Nieqing Cao , Jin Wang , Chao Yu , Yan Ding
PDF Cite Dataset Project
UMI-Bench 1.0: An Open and Reproducible Real-World Benchmark for Tabletop Robotic Manipulation with UMI Data
GN0: Toward a Unified Paradigm for Generation, Evaluation, and Policy Learning in Visual-Language Navigation
arXiv preprint
GN0 unifies GN-Matrix data generation, the high-fidelity GN-Bench simulator, and the RL-driven GN-BAE foundation model for map-based and map-free visual-language navigation.
Xinhai Li , Xiaotao Zhang , Yuehao Huang , Jiankun Dong , Tianhang Wang , Sunyao Zhou , Yunzi Wu , Chengnuo Sun , Yunfei Ge , Qizhen Weng , Chi Zhang , Chenjia Bai✉ , Xuelong Li✉
PDF Cite Code Project Models
GN0: Toward a Unified Paradigm for Generation, Evaluation, and Policy Learning in Visual-Language Navigation
WNM-3D: A World Navigation Model with 3D Scene Conditioning for Closed-Loop VLN
arXiv preprint, 2026
WNM-3D conditions joint future-view and action generation on persistent 3D scene tokens, combining imitation learning, DAgger, and reinforcement learning for closed-loop visual-language navigation.
Yuehao Huang✉ , Yunzi Wu , Xiaotao Zhang , Xinhai Li , Jiankun Dong , Jiajun Lv , Chi Zhang , Chenjia Bai✉ , Yong Liu , Xuelong Li
PDF Cite Code Project Models
WNM-3D: A World Navigation Model with 3D Scene Conditioning for Closed-Loop VLN
ReWeight: Leveraging Human Data for VLA Post-Training via Demonstration Retrieval and Sample Weighting
arXiv preprint, 2026
ReWeight retrieves relevant human demonstrations and weights individual samples through cross-embodiment visuomotor matching, improving VLA post-training in simulation and real-world manipulation.
Chenwei Wang , Dianye Huang , Match W.L. Ko , Chenjia Bai , Zhongliang Jiang✉
PDF Cite Project
ReWeight: Leveraging Human Data for VLA Post-Training via Demonstration Retrieval and Sample Weighting
V-JEPA Policy: Building Effective World-Action Models on Predictive Visual Latents
arXiv preprint, 2026
V-JEPA Policy builds a compact world-action model on frozen predictive visual latents, jointly learning future prediction and action generation while transferring knowledge from videos without action labels.
Yang Zhang , Jiangyuan Zhao , Chenyou Fan , Jiayu Hu , Xiu Yuan , Chenjia Bai , Xiu Li✉
PDF Cite Code
V-JEPA Policy: Building Effective World-Action Models on Predictive Visual Latents
NEXUS: Perceptive Whole-Body Control for Terrain-Adaptive Teleoperation
arXiv preprint, 2026
NEXUS combines scalable terrain-aware motion adaptation and perceptive whole-body control to reproduce human motion commands across uneven terrains, enabling zero-shot real-world teleoperation.
Xiangyu Miao , Junsong Wu , Jiyuan Shi , Weiji Xie , Jinrui Han , Xingyi Wang , Weinan Zhang✉ , Chenjia Bai✉ , Xuelong Li
PDF Cite Project
NEXUS: Perceptive Whole-Body Control for Terrain-Adaptive Teleoperation