Beyond MACs: Hardware Efficient Architecture Design for Vision Backbones

ArXi:2603.26551v1 Announce Type: cross Vision backbone networks play a central role in modern computer vision. Enhancing their efficiency directly benefits a wide range of downstream applications. To measure efficiency, many publications rely on MACs (Multiply Accumulate operations) as a predictor of execution time. In this paper, we experimentally nstrate the shortcomings of such a metric, especially in the context of edge devices.