QuantVLA: Scale-Calibrated Post-Training Quantization for Vision-Language-Action Models Paper • 2602.20309 • Published 3 days ago • 10