Home /Breakthroughs /Gemini Robotics 1.5 enables cross-embodiment skill transfer without specialization
🧪 Technology⚡· Impact 7/10
Gemini Robotics 1.5 enables cross-embodiment skill transfer without specialization
Source: deepmind.google · May 25, 2026
Summary
Google DeepMind's Gemini Robotics 1.5 allows a robot to transfer learned motions to another robot without task-specific retraining, significantly accelerating learning across different physical forms. This is a first-of-its-kind capability that bridges the gap between simulation and diverse real-world embodiments.
Related research
All papers →Recent papers matching 4 keywords from this headline
MANIPULATION
Open access
Instrumentation for Imitation Learning: Enhancing Training Datasets for Clothes Hanger Insertion
Remko Proesmans, Thomas Lips, Francis wyffels
2026
LEARNING
Open access
Direct Dynamic Retargeting for Humanoid Imitation Learning from Videos
Constant Roux, Ludovic De Matteïs, Armand Jordana +4 more
2026
LEARNING
Open access
Any2Any: Efficient Cross-Embodiment Transfer for Humanoid Whole-Body Tracking
Ming Yang, Tao Yu, Feng Li +1 more
2026
MANIPULATION
Open access
Semantically Structured Mixture-of-Experts for Compositional Robotic Manipulation
Chengyu Deng, Guanqi Chen, Yizhou Chen +4 more
2026