DocMamba:利用状态空间模型进行高效文档预培训

Pengfei Hu, Zhenrong Zhang, Jiefeng Ma, Shuhang Liu, Jun Du, Jianshu Zhang
{"title":"DocMamba:利用状态空间模型进行高效文档预培训","authors":"Pengfei Hu, Zhenrong Zhang, Jiefeng Ma, Shuhang Liu, Jun Du, Jianshu Zhang","doi":"arxiv-2409.11887","DOIUrl":null,"url":null,"abstract":"In recent years, visually-rich document understanding has attracted\nincreasing attention. Transformer-based pre-trained models have become the\nmainstream approach, yielding significant performance gains in this field.\nHowever, the self-attention mechanism's quadratic computational complexity\nhinders their efficiency and ability to process long documents. In this paper,\nwe present DocMamba, a novel framework based on the state space model. It is\ndesigned to reduce computational complexity to linear while preserving global\nmodeling capabilities. To further enhance its effectiveness in document\nprocessing, we introduce the Segment-First Bidirectional Scan (SFBS) to capture\ncontiguous semantic information. Experimental results demonstrate that DocMamba\nachieves new state-of-the-art results on downstream datasets such as FUNSD,\nCORD, and SORIE, while significantly improving speed and reducing memory usage.\nNotably, experiments on the HRDoc confirm DocMamba's potential for length\nextrapolation. The code will be available online.","PeriodicalId":501030,"journal":{"name":"arXiv - CS - Computation and Language","volume":"30 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"DocMamba: Efficient Document Pre-training with State Space Model\",\"authors\":\"Pengfei Hu, Zhenrong Zhang, Jiefeng Ma, Shuhang Liu, Jun Du, Jianshu Zhang\",\"doi\":\"arxiv-2409.11887\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In recent years, visually-rich document understanding has attracted\\nincreasing attention. Transformer-based pre-trained models have become the\\nmainstream approach, yielding significant performance gains in this field.\\nHowever, the self-attention mechanism's quadratic computational complexity\\nhinders their efficiency and ability to process long documents. In this paper,\\nwe present DocMamba, a novel framework based on the state space model. It is\\ndesigned to reduce computational complexity to linear while preserving global\\nmodeling capabilities. To further enhance its effectiveness in document\\nprocessing, we introduce the Segment-First Bidirectional Scan (SFBS) to capture\\ncontiguous semantic information. Experimental results demonstrate that DocMamba\\nachieves new state-of-the-art results on downstream datasets such as FUNSD,\\nCORD, and SORIE, while significantly improving speed and reducing memory usage.\\nNotably, experiments on the HRDoc confirm DocMamba's potential for length\\nextrapolation. The code will be available online.\",\"PeriodicalId\":501030,\"journal\":{\"name\":\"arXiv - CS - Computation and Language\",\"volume\":\"30 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-18\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - CS - Computation and Language\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.11887\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Computation and Language","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.11887","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

近年来,视觉丰富的文档理解吸引了越来越多的关注。然而,自注意机制的二次计算复杂性阻碍了其处理长文档的效率和能力。在本文中,我们介绍了基于状态空间模型的新型框架 DocMamba。它旨在将计算复杂度降至线性,同时保留全局建模能力。为了进一步提高其在文档处理中的有效性,我们引入了分段优先双向扫描(SFBS)来捕捉连续的语义信息。实验结果表明,DocMamba在FUNSD、CORD和SORIE等下游数据集上取得了新的一流成果,同时显著提高了速度并减少了内存使用。代码将在网上公布。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
DocMamba: Efficient Document Pre-training with State Space Model
In recent years, visually-rich document understanding has attracted increasing attention. Transformer-based pre-trained models have become the mainstream approach, yielding significant performance gains in this field. However, the self-attention mechanism's quadratic computational complexity hinders their efficiency and ability to process long documents. In this paper, we present DocMamba, a novel framework based on the state space model. It is designed to reduce computational complexity to linear while preserving global modeling capabilities. To further enhance its effectiveness in document processing, we introduce the Segment-First Bidirectional Scan (SFBS) to capture contiguous semantic information. Experimental results demonstrate that DocMamba achieves new state-of-the-art results on downstream datasets such as FUNSD, CORD, and SORIE, while significantly improving speed and reducing memory usage. Notably, experiments on the HRDoc confirm DocMamba's potential for length extrapolation. The code will be available online.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信