BaSFormer: A Balanced Sparsity Regularized Attention Network for Transformer

Author:

Jiang Shuoran1ORCID,Chen Qingcai1ORCID,Xiang Yang2ORCID,Pan Youcheng2ORCID,Wu Xiangping1ORCID

Affiliation:

1. Harbin Institute of Technology, Shenzhen, China

2. Peng Cheng Laboratory, Shenzhen, China

Funder

National Key R&D Program of China

Shenzhen Science and Technology Plan

Science and Technology Department of Guizhou Province

National Social Science Fund of China

China Postdoctoral Science Foundation

Development and Reform Commission of Shenzhen Municipality

National Natural Science Foundation of China

Guangdong Provincial Key Laboratory

PCL

Key Laboratory of Intelligent Computing in Network Environment

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Reference98 articles.

1. A survey of large language models;Zhao,2023

2. Chat Generative Pre-trained Transformer: why we should embrace this technology

3. Exploring the limits of transfer learning with a unified text-to-text transformer;Raffel;J. Mach. Learn. Res.,2020

4. LLaMA: Open and efficient foundation language models;Touvron,2023

5. Obtaining genetics insights from deep learning via explainable artificial intelligence

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