arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2401.09220cs.CL
收藏

UniVIE:一种面向表单类文档视觉信息提取的统一标签空间方法

UniVIE: A Unified Label Space Approach to Visual Information Extraction from Form-like Documents

University of Science and Technology of China(中国科学技术大学);Microsoft Research Asia(微软亚洲研究院)

Kai Hu, Jiawei Wang, Weihong Lin, Zhuoyao Zhong, Lei Sun, Qiang Huo

AI总结:

UniVIE 将表单视觉信息提取统一为关系预测问题,通过单一标签空间和由粗到细的树生成策略,结合树注意力掩码与树层级嵌入,在 HierForms 和 SIBR 数据集上取得最先进结果。

摘要:

现有的表单类文档视觉信息提取(VIE)方法通常将流程拆分为多个独立子任务,例如关键信息提取、键值对提取和选项组提取。然而,这些方法往往忽略了表单文档的层次结构,包括层次化键值对和层次化选项组。为了解决这些局限性,我们提出了一种新的视角,将 VIE 重新定义为关系预测问题,并将不同任务的标签统一到单一标签空间中。这种统一方法能够定义多种关系类型,并有效处理表单类文档中的层次关系。基于这一视角,我们提出了 UniVIE,这是一个全面解决 VIE 问题的统一模型。UniVIE 采用由粗到细的策略运行。它首先通过树提议网络生成树提议,随后由关系解码器模块将其细化为层次树。为了增强 UniVIE 的关系预测能力,我们在关系解码器中引入了两种新颖的树约束:树注意力掩码和树层级嵌入。在我们自建的 HierForms 数据集和公开可用的 SIBR 数据集上进行了大量实验评估,结果证实我们的方法取得了最先进的结果,凸显了我们的统一方法在推动 VIE 领域发展方面的有效性和潜力。

英文摘要:

Existing methods for Visual Information Extraction (VIE) from form-like documents typically fragment the process into separate subtasks, such as key information extraction, key-value pair extraction, and choice group extraction. However, these approaches often overlook the hierarchical structure of form documents, including hierarchical key-value pairs and hierarchical choice groups. To address these limitations, we present a new perspective, reframing VIE as a relation prediction problem and unifying labels of different tasks into a single label space. This unified approach allows for the definition of various relation types and effectively tackles hierarchical relationships in form-like documents. In line with this perspective, we present UniVIE, a unified model that addresses the VIE problem comprehensively. UniVIE functions using a coarse-to-fine strategy. It initially generates tree proposals through a tree proposal network, which are subsequently refined into hierarchical trees by a relation decoder module. To enhance the relation prediction capabilities of UniVIE, we incorporate two novel tree constraints into the relation decoder: a tree attention mask and a tree level embedding. Extensive experimental evaluations on both our in-house dataset HierForms and a publicly available dataset SIBR, substantiate that our method achieves state-of-the-art results, underscoring the effectiveness and potential of our unified approach in advancing the field of VIE.

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。

↑