New AI Robots Learn Twice as Fast by Knowing Their Own Bodies

Researchers tested GPT-6-Astra controlling an XLeRobot in a simulated and a physical elevator-button task. In 30 fixed-start simulation trials, complete robot geometry and camera information reduced mean completion time by 57.4% relative to a baseline with only the common control interface and no prior experience; images with synchronized action and state records reduced it by 68.6% without additional body assets. In nine paired comparisons (18 trials) at starts displaced by 10–100 cm, experience recorded at the original start reduced mean time by 58–63% relative to no experience, demonstrating generalization to the tested new starting positions. During experience experiments, GPT-6-Astra spontaneously generated a short visual-feedback program, and researcher-refactored versions reduced mean local-task time by 29–31% in 27 simulation trials. Finally, 12 real-robot trials using operator-confirmed button contact demonstrated sim2real reuse: at a shared nominal start, simulation XML assets and simulation experience reduced mean time by 53.0% and 49.9% respectively, and real experience also transferred to two new starts. The authors suggest a practical way to build general-purpose manipulation experiments around GPT-6-Astra: supply machine-readable body descriptions and synchronized demonstrations, and turn useful agent-generated feedback routines into reusable skills, while the agent adapts actions from current images.