Actra: Optimized Transformer Architecture for Vision-Language-Action Models in Robot Learning

Re-assign community