Microsoft Extends Its Phi Models To Physical AI With Rho-Alpha

Physical AI marks a transition from robots as programmed tools to robots as adaptable collaborators. That transition will ...
IEEE

DexSkills: Skill Segmentation Using Haptic Data for Learning Autonomous Long-Horizon Robotic Manipulation Tasks

Abstract: Effective execution of long-horizon tasks with dexterous robotic hands remains a significant challenge in real-world problems. While learning from human demonstrations has shown encouraging ...
IEEE

Gram: A Large-Scale General EEG Model for Raw Data Classification and Restoration Tasks

Abstract: Drawing insights from Large Language Models, researchers have developed several large-scale Electroencephalogram (EEG) models (LEMs) to learn a generalized representation adaptable to ...
Microsoft

TwinVLA: Data-Efficient Bimanual Manipulation with Twin Single-Arm Vision-Language-Action Models

Vision-language-action models (VLAs) trained on large-scale robotic datasets have demonstrated strong performance on manipulation tasks, including bimanual tasks. However, because most public datasets ...