13 papers
Training in Deployment: Human-in-the-loop Online Post-Training for Robot Manipulation
Chun Gu, Xiaofei Wei, Yiming Chen +4
Large pretrained robot foundation models provide strong priors for manipulation, yet adapting them to target tasks with offline teleoperated demonstrations often fails to cover the…
SMR: Learning streaming motion retargeting for humanoids from human semantic
Chun Gu, Juntong Zhang, Jingyu Li +1
Retargeting human motion to humanoids requires preserving motion semantics while accounting for the robot's physical constraints. Kinematic pipelines prescribe a robot trajectory t…
Forward Dynamics Models
Zipei Ma, Junzhe Jiang, Jiahui Zhang +6
Vision-language-action (VLA) models and world action models (WAMs) have advanced robotic policy learning by building on pretrained vision-language models (VLMs) and video generativ…
PRICE the Action Chunks: Physical Relational Credit Assignment for Embodied Reinforcement Learning
Yangang Zou, Jiajun Lu, Weitao Zhou +6
Outcome-based reinforcement learning (RL) post-trains vision--language--action policies using terminal success signals, but assigns the same trajectory-level advantage to every act…
X-WBC: A Cross-Embodiment Foundation Model for Humanoid Whole-Body Control
Juntong Zhang, Chun Gu, Li Zhang
Scaling humanoid whole-body control toward general-purpose deployment requires large human motion corpora and training experience shared across robot bodies. Existing methods usual…
Learning Transferable Dynamics Priors from Action to World Modeling
Ze Huang, Jiahui Zhang, Hairuo Liu +3
We study action-conditioned world modeling as a scalable way to learn transferable dynamics priors for robot learning. By pretraining a model to predict how actions drive visual sc…
ROVE: Unlocking Human Interventions for Humanoid Manipulation via Reinforcement Learning
Wei Xiao, Weiliang Tang, Yuying Ge +4
Human interventions provide crucial corrective signals for post-training Vision-Language-Action (VLA) models. However, enabling seamless humanoid interventions is a formidable syst…
Metis: A Generalizable and Efficient World-Action Model for Autonomous Driving and Urban Navigation
Jingyu Li, Zhe Liu, Dongnan Hu +10
World action models~(WAMs) have shown great promise for autonomous driving and urban navigation. Built upon Vision-Language-Action models or video generation models, existing appro…
RoboRouter: Training-Free Policy Routing for Robotic Manipulation
Yiteng Chen, Zhe Cao, Hongjia Ren +9
Research on robotic manipulation has developed a diverse set of policy paradigms, including vision-language-action (VLA) models, vision-action (VA) policies, and code-based composi…
Reinforcing Action Policies by Prophesying
Jiahui Zhang, Ze Huang, Chun Gu +2
Vision-Language-Action (VLA) policies excel in aligning language, perception, and robot control. However, most VLAs are trained purely by imitation, which overfits to demonstration…
UniUGG: Unified 3D Understanding and Generation via Geometric-Semantic Encoding
Yueming Xu, Jiahui Zhang, Ze Huang +12
Despite the impressive progress on understanding and generating images shown by the recent unified architectures, the integration of 3D tasks remains challenging and largely unexpl…
BézierGS: Dynamic Urban Scene Reconstruction with Bézier Curve Gaussian Splatting
Zipei Ma, Junzhe Jiang, Yurui Chen +1
The realistic reconstruction of street scenes is critical for developing real-world simulators in autonomous driving. Most existing methods rely on object pose annotations, using t…
4D-VLA: Spatiotemporal Vision-Language-Action Pretraining with Cross-Scene Calibration
Jiahui Zhang, Yurui Chen, Yueming Xu +8
Leveraging diverse robotic data for pretraining remains a critical challenge. Existing methods typically model the dataset's action distribution using simple observations as inputs…