AI RESEARCH

OpenACMv2: An Accuracy-Constrained Co-Optimization Framework for Approximate DCiM

arXiv CS.LG

ArXi:2603.13042v1 Announce Type: new Digital Compute-in-Memory (DCiM) accelerates neural networks by reducing data movement. Approximate DCiM can further improve power-performance-area (PPA), but demands accuracy-constrained co-optimization across coupled architecture and transistor-level choices. Building on OpenYield, we