6 papers
Cosmos 3: Omnimodal World Models for Physical AI
NVIDIA, :, Aditi +293
We introduce Cosmos 3, a family of omnimodal world models designed to jointly process and generate language, image, video, audio, and action sequences within a unified mixture-of-t…
Learning Dexterous Manipulation Using Contact Wrench Guidance From Human Demonstration
Xinghao Zhu, Zixi Liu, Shalin Jain +18
Dexterous robot manipulation can benefit from the abundance of human demonstrations, but transferring such demonstrations to robot policies remains challenging. We present Contact…
AGILE: A Comprehensive Workflow for Humanoid Loco-Manipulation Learning
Huihua Zhao, Rafael Cathomen, Lionel Gulich +6
Recent advances in reinforcement learning (RL) have enabled impressive humanoid behaviors in simulation, yet transferring these results to new robots remains challenging. In many r…
Isaac Lab: A GPU-Accelerated Simulation Framework for Multi-Modal Robot Learning
NVIDIA, :, Mayank Mittal +104
We present Isaac Lab, the natural successor to Isaac Gym, which extends the paradigm of GPU-native robotics simulation into the era of large-scale multi-modal learning. Isaac Lab c…
COMPASS: Cross-embodiment Mobility Policy via Residual RL and Skill Synthesis
Wei Liu, Huihua Zhao, Chenran Li +4
As robots are increasingly deployed in diverse application domains, enabling robust mobility across different embodiments has become a critical challenge. Classical mobility stacks…
X-MOBILITY: End-To-End Generalizable Navigation via World Modeling
Wei Liu, Huihua Zhao, Chenran Li +5
General-purpose navigation in challenging environments remains a significant problem in robotics, with current state-of-the-art approaches facing myriad limitations. Classical appr…