robotics research

Microsoft Research has presented an object-centric residual reinforcement learning method that trains a lightweight corrective robot policy entirely in simulation, adds it to a frozen vision-language-action model, and reports zero-shot real-robot gains across five manipulation tasks from 42...

Nvidia and academic collaborators on June 17, 2026, detailed ENPIRE, a robotics research framework that lets AI coding agents run real hardware experiments, verify results, rewrite policy code, and iterate toward tasks such as installing a GPU into a motherboard. The headline image is...