Enhancing Long Sequence Input Processing in FPGA-Based Transformer Accelerators through Attention Fusion

Author:

Qin Yunji1ORCID,Lou Wenqi2ORCID,Wang Chao1ORCID,Gong Lei1ORCID,Zhou Xuehai1ORCID

Affiliation:

1. School of Computer Science and Technology, University of Science and Technology of China, China

2. Software Engineering, University of Science and Technology of China, China

Funder

Jiangsu Provincial Natural Science Foundation

Youth Innovation Promotion Association of the Chinese Academy of Sciences

National Key Research and Development Program of China

National Natural Science Foundation of China

Publisher

ACM

Reference20 articles.

1. Tri Dao 2022. Flashattention: Fast and memory-efficient exact attention with io-awareness. In NeurIPS. 16344–16359.

2. Jacob Devlin 2018. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805 (2018).

3. Alexey Dosovitskiy 2020. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929 (2020).

4. Adaptable Butterfly Accelerator for Attention-based NNs via Hardware and Algorithm Co-design

5. An algorithm–hardware co-optimized framework for accelerating n: M sparse transformers;Fang Chao;TVLSI,2022

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3