11 papers
Flow Matching for Robot Foundation Models
Kuangji Zuo, Tuo An, Xinying Guo +9
Flow matching is a common generative recipe for robot foundation models. Vision-language-action models (VLAs) generate action chunks, while world-action models (WAMs) additionally…
Unified Condition-Action Modeling for Accurate One-Step Action Generation
Xinyu Zhou, Zikun Cai, Kuangji Zuo +7
Robot manipulation requires policies that are both accurate and efficient, as robot control must respond to changing observations under tight latency constraints. Recent diffusion…
USS: Unifying Spatial-Semantic Prompting for End to End Embodied Visual Tracking
Yuchen Xie, Xinyu Zhou, Kuangji Zuo +4
Embodied Visual Tracking (EVT) requires an agent to continuously follow a designated target while moving through dynamic environments. Existing embodied tracking methods generally…
ADAPT: Analytical Disturbance-Aware Policy Training for Humanoid Locomotion
Bofan Lyu, Jindou Jia, Kuangji Zuo +7
Humanoids deployed in human-centered environments must handle force-interactive tasks, where external contacts introduce unexpected disturbances that disrupt locomotion accuracy an…
EM-Fall: Embodied mmWave Sensing for Day-and-Night Fall Detection on Humanoid Robots
Yanshuo Lu, Yuxuan Hu, Shenghai Yuan +5
Falls are one of the leading causes of injury and hospitalization among elderly individuals, making reliable fall awareness an essential capability for safety monitoring in residen…
Gaze2Act: Gaze-Conditioned Vision-Language-Action Policies for Interactive Robot Manipulation
Kuangji Zuo, Gen Li, Bofan Lyu +9
Vision-Language-Action (VLA) models have recently shown strong potential for robot learning by following language instructions. However, in practice, language alone is often insuff…
OccamToken: Efficient VLM Inference with Training-Free and Budget-Adaptive Token Pruning
Geng Li, Guohao Chen, Ting Chen +6
Vision-language models (VLMs) rely on long visual token sequences for visual understanding, making the prefill stage expensive in both computation and memory. Most existing pruning…
FLASH: Efficient Visuomotor Policy via Sparse Sampling
Jiaqi Bai, Jindou Jia, Yuxuan Hu +5
Generative models such as diffusion and flow matching have become dominant paradigms for visuomotor policy learning, yet their reliance on iterative denoising incurs high inference…
RoboForge: Physically Optimized Text-guided Whole-Body Locomotion for Humanoids
Xichen Yuan, Zhe Li, Bofan Lyu +4
While generative models have become effective at producing human-like motions from text, transferring these motions to humanoid robots for physical execution remains challenging. E…
Real-Time Human Activity Recognition on Edge Microcontrollers: Dynamic Hierarchical Inference with Multi-Spectral Sensor Fusion
Boyu Li, Kuangji Zuo, Lincong Li +1
The demand for accurate on-device pattern recognition in edge applications is intensifying, yet existing approaches struggle to reconcile accuracy with computational constraints. T…
WaveMan: mmWave-Based Room-Scale Human Interaction Perception for Humanoid Robots
Yuxuan Hu, Kuangji Zuo, Boyu Ma +4
Reliable humanoid-robot interaction (HRI) in household environments is constrained by two fundamental requirements, namely robustness to unconstrained user positions and preservati…