arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2512.07855cs.LGcs.CV
收藏

LAPA: 一种基于对数域的预测驱动型动态稀疏加速器用于Transformer模型

LAPA: Log-Domain Prediction-Driven Dynamic Sparsity Accelerator for Transformer Model

School of Integrated Circuits Tsinghua University Beijing, China(集成电路学院清华大学北京中国);School of Integrated Circuits Tsinghua University Beijing, China Shanghai Artificial Intelligence Laboratory Shanghai, China(集成电路学院清华大学北京中国上海人工智能实验室上海中国)

Huizheng Wang, Hongbin Wang, Shaojun Wei, Yang Hu, Shouyi Yin

AI总结:

LAPA通过设计不对称领先一计算方案和混合精度多轮位移累积机制,提升Transformer模型的计算效率和能效。

摘要:

基于注意力机制的Transformer模型已彻底改变了自然语言处理(NLP)领域,并在计算机视觉(CV)任务中表现出强大的性能。然而,随着输入序列的变化,Transformer模型中的计算瓶颈在各个阶段表现出动态行为,这要求一种跨阶段的稀疏加速策略。不幸的是,大多数现有的稀疏Transformer方法都是单阶段的,其稀疏性预测机制在应用于多个阶段时会导致显著的功耗开销。为此,本文提出了一种基于对数域的注意力预测算法-架构协同设计,命名为LAPA。首先,设计了一种不对称领先一计算(ALOC)方案以消除昂贵的乘法运算。其次,进一步提出了一种混合精度多轮位移累积(MRSA)机制以缓解累积开销。设计了一种数据特征依赖的过滤器(DDF)策略以与MRSA过程协同工作。最后,设计了一种精心设计的加速器,将理论上的提升转化为实际的硬件改进。实验结果表明,LAPA在能量效率方面分别比最先进的(SOTA)工作Spatten、Sanger和FACT高出3.52倍、3.24倍和2.79倍。

英文摘要:

Attention-based Transformers have revolutionized natural language processing (NLP) and shown strong performance in computer vision (CV) tasks. However, as the input sequence varies, the computational bottlenecks in Transformer models exhibit dynamic behavior across stages, which calls for a cross-stage sparse acceleration strategy. Unfortunately, most existing sparse Transformer approaches are single-stage based, and their sparsity prediction mechanisms lead to significant power overhead when applied across multiple stages. To this end, this paper proposes a log-domain attention prediction algorithm-architecture co-design, named LAPA. First, an asymmetric leading one computing (ALOC) scheme is designed to eliminate expensive multiplications. Next, a mixed-precision multi-round shifting accumulation (MRSA) mechanism is further proposed to mitigate the accumulation overhead. A data-feature dependent filter (DDF) strategy is designed to work in concert with the MRSA process. Finally, an elaborate accelerator is designed to translate the theoretical enhancement into practical hardware improvement. Experimental results show that LAPA achieves 3.52x, 3.24x and 2.79x higher energy efficiency than the state-of-the-art (SOTA) works Spatten, Sanger and FACT, respectively.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑