MGDoc:用于文档图像理解的多粒度层次预训练

Zilong Wang, Jiuxiang Gu, Chris Tensmeyer, Nikolaos Barmpalios, A. Nenkova, Tong Sun, Jingbo Shang, Vlad I. Morariu
{"title":"MGDoc:用于文档图像理解的多粒度层次预训练","authors":"Zilong Wang, Jiuxiang Gu, Chris Tensmeyer, Nikolaos Barmpalios, A. Nenkova, Tong Sun, Jingbo Shang, Vlad I. Morariu","doi":"10.48550/arXiv.2211.14958","DOIUrl":null,"url":null,"abstract":"Document images are a ubiquitous source of data where the text is organized in a complex hierarchical structure ranging from fine granularity (e.g., words), medium granularity (e.g., regions such as paragraphs or figures), to coarse granularity (e.g., the whole page). The spatial hierarchical relationships between content at different levels of granularity are crucial for document image understanding tasks. Existing methods learn features from either word-level or region-level but fail to consider both simultaneously. Word-level models are restricted by the fact that they originate from pure-text language models, which only encode the word-level context. In contrast, region-level models attempt to encode regions corresponding to paragraphs or text blocks into a single embedding, but they perform worse with additional word-level features. To deal with these issues, we propose MGDoc, a new multi-modal multi-granular pre-training framework that encodes page-level, region-level, and word-level information at the same time. MGDoc uses a unified text-visual encoder to obtain multi-modal features across different granularities, which makes it possible to project the multi-granular features into the same hyperspace. To model the region-word correlation, we design a cross-granular attention mechanism and specific pre-training tasks for our model to reinforce the model of learning the hierarchy between regions and words. Experiments demonstrate that our proposed model can learn better features that perform well across granularities and lead to improvements in downstream tasks.","PeriodicalId":74540,"journal":{"name":"Proceedings of the Conference on Empirical Methods in Natural Language Processing. Conference on Empirical Methods in Natural Language Processing","volume":"12 1","pages":"3984-3993"},"PeriodicalIF":0.0000,"publicationDate":"2022-11-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":"{\"title\":\"MGDoc: Pre-training with Multi-granular Hierarchy for Document Image Understanding\",\"authors\":\"Zilong Wang, Jiuxiang Gu, Chris Tensmeyer, Nikolaos Barmpalios, A. Nenkova, Tong Sun, Jingbo Shang, Vlad I. Morariu\",\"doi\":\"10.48550/arXiv.2211.14958\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Document images are a ubiquitous source of data where the text is organized in a complex hierarchical structure ranging from fine granularity (e.g., words), medium granularity (e.g., regions such as paragraphs or figures), to coarse granularity (e.g., the whole page). The spatial hierarchical relationships between content at different levels of granularity are crucial for document image understanding tasks. Existing methods learn features from either word-level or region-level but fail to consider both simultaneously. Word-level models are restricted by the fact that they originate from pure-text language models, which only encode the word-level context. In contrast, region-level models attempt to encode regions corresponding to paragraphs or text blocks into a single embedding, but they perform worse with additional word-level features. To deal with these issues, we propose MGDoc, a new multi-modal multi-granular pre-training framework that encodes page-level, region-level, and word-level information at the same time. MGDoc uses a unified text-visual encoder to obtain multi-modal features across different granularities, which makes it possible to project the multi-granular features into the same hyperspace. To model the region-word correlation, we design a cross-granular attention mechanism and specific pre-training tasks for our model to reinforce the model of learning the hierarchy between regions and words. Experiments demonstrate that our proposed model can learn better features that perform well across granularities and lead to improvements in downstream tasks.\",\"PeriodicalId\":74540,\"journal\":{\"name\":\"Proceedings of the Conference on Empirical Methods in Natural Language Processing. Conference on Empirical Methods in Natural Language Processing\",\"volume\":\"12 1\",\"pages\":\"3984-3993\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-11-27\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"6\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of the Conference on Empirical Methods in Natural Language Processing. Conference on Empirical Methods in Natural Language Processing\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.48550/arXiv.2211.14958\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the Conference on Empirical Methods in Natural Language Processing. Conference on Empirical Methods in Natural Language Processing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.48550/arXiv.2211.14958","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 6

摘要

文档图像是一种无处不在的数据源,其中文本以复杂的层次结构组织,范围从细粒度(例如,单词),中等粒度(例如,段落或图形等区域)到粗粒度(例如,整个页面)。不同粒度级别的内容之间的空间层次关系对于文档图像理解任务至关重要。现有的方法要么从词级学习特征,要么从区域级学习特征,但没有同时考虑这两个方面。词级模型受到纯文本语言模型的限制,纯文本语言模型只对词级上下文进行编码。相比之下,区域级模型试图将对应于段落或文本块的区域编码到单个嵌入中,但它们在附加词级特征时表现较差。为了解决这些问题,我们提出了一种新的多模态多粒度预训练框架MGDoc,它可以同时编码页面级、区域级和词级信息。MGDoc使用统一的文本-视觉编码器获得不同粒度的多模态特征,这使得将多粒度特征投影到同一超空间成为可能。为了建立区域-词相关的模型,我们设计了一个跨粒度的注意机制和特定的预训练任务,以加强区域和词之间的层次学习模型。实验表明,我们提出的模型可以学习更好的特征,这些特征在粒度上表现良好,并导致下游任务的改进。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
MGDoc: Pre-training with Multi-granular Hierarchy for Document Image Understanding
Document images are a ubiquitous source of data where the text is organized in a complex hierarchical structure ranging from fine granularity (e.g., words), medium granularity (e.g., regions such as paragraphs or figures), to coarse granularity (e.g., the whole page). The spatial hierarchical relationships between content at different levels of granularity are crucial for document image understanding tasks. Existing methods learn features from either word-level or region-level but fail to consider both simultaneously. Word-level models are restricted by the fact that they originate from pure-text language models, which only encode the word-level context. In contrast, region-level models attempt to encode regions corresponding to paragraphs or text blocks into a single embedding, but they perform worse with additional word-level features. To deal with these issues, we propose MGDoc, a new multi-modal multi-granular pre-training framework that encodes page-level, region-level, and word-level information at the same time. MGDoc uses a unified text-visual encoder to obtain multi-modal features across different granularities, which makes it possible to project the multi-granular features into the same hyperspace. To model the region-word correlation, we design a cross-granular attention mechanism and specific pre-training tasks for our model to reinforce the model of learning the hierarchy between regions and words. Experiments demonstrate that our proposed model can learn better features that perform well across granularities and lead to improvements in downstream tasks.
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
期刊最新文献
EHRAgent: Code Empowers Large Language Models for Few-shot Complex Tabular Reasoning on Electronic Health Records. Two Directions for Clinical Data Generation with Large Language Models: Data-to-Label and Label-to-Data. Hierarchical Pretraining on Multimodal Electronic Health Records. An Integrative Survey on Mental Health Conversational Agents to Bridge Computer Science and Medical Perspectives. A Comprehensive Evaluation of Biomedical Entity Linking Models.
×
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
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 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 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1