Recent research efforts focus on reducing the computational and memory overheads of Large Language Models (LLMs) to make them feasible on resource-constrained devices. Despite advancements in compression techniques, non-linear operators like Softmax and Layernorm remain bottlenecks due to their sensitivity to quantization. We propose SoftmAP, a software-hardware co-design methodology that implements an integer-only low-precision Softmax using In-Memory Compute (IMC) hardware. Our method achieves up to three orders of magnitude improvement in the energy-delay product compared to A100 and RTX3090 GPUs, making LLMs more deployable without compromising performance.
@article{arxiv.2411.17847,
title = {SoftmAP: Software-Hardware Co-design for Integer-Only Softmax on Associative Processors},
author = {Mariam Rakka and Jinhao Li and Guohao Dai and Ahmed Eltawil and Mohammed E. Fouda and Fadi Kurdahi},
journal= {arXiv preprint arXiv:2411.17847},
year = {2024}
}