DoT: An efficient Double Transformer for NLP tasks with tables

Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, pages 330–343 August 1–6, 2021. ©2021 Association for Computational Linguistics 330 Spatial Dependency Parsing for Semi-Structured Document Information Extraction Wonseok Hwang Jinyeong Yim Seunghyun Park Sohee Yang Minjoon Seo Clova AI, NAVER Corp. {wonseok.hwang,jinyeong.yim,seung.park,sh.yang,minjoon.seo} @navercorp.com Abstract Information Extraction (IE) for semi- structured document images is often ap- proached as a sequence tagging problem by classifying each recognized input token into one of the IOB (Inside, Outside, and Beginning) categories. However, such prob- lem setup has two inherent limitations that (1) it cannot easily handle complex spatial relationships and (2) it is not suitable for highly structured information, which are nevertheless frequently observed in real-world document images. To tackle these issues, we first formulate the IE task as spatial dependency parsing problem that focuses on the relationship among text tokens in the documents. Under this setup, we then propose SPADE� (SPAtial DEpendency parser) that models highly complex spatial relationships and an arbitrary number of information layers in the documents in an end-to-end manner. We evaluate it on various kinds of documents such as receipts, name cards, forms, and invoices, and show that it achieves a similar or better performance compared to strong baselines including BERT-based IOB taggger. 1 Introduction Document information extraction (IE) is the task of mapping each document to a structured form that is consistent with the target ontology (e.g., database schema), which has become an increas- ingly important task in both research community and industry. In this paper, we are particularly in- terested in information extraction from real-world, semi-structured document images, such as invoices, receipts, and name cards, where we assume Opti- cal Character Recognition (OCR, i.e. detecting the locations of the text tokens if the input is an image) has been already applied. Previous approaches for semi-structured document IE often assume as if the input is a one-dimensional sequence and formu- late the task as an IOB (Inside Outside Beginning) tagging problem. In this setup, the tokens in the document (either obtained through an OCR engine or trivially parsed from a web page or pdf) are first serialized, and then an independent tagging model classifies each of the flattened lists into one of the pre-defined IOB categories (Ramshaw and Mar- cus, 1995; Palm et al., 2017). While effective for relatively simple documents, their broader applica- tion in the real world is still challenging because (1) semi-structured documents often exhibit a com- plex layout where the serialization algorithm is non-trivial, and (2) sequence tagging is inherently not effective for encoding multi-layer hierarchi- cal information such as the menu tree in receipts (Fig. 1c). To overcome these limitations, we propose SPADE� (SPAtial DEpendency parser), an end-to- end, serializer-free model that is capable of extract- ing hierarchical information from complex docu- ments. Rather than explicitly dividing the original problem into two independent subtasks of serializa- tion and tagging, our model tackles the problem in an end-to-end manner by creating a directed rela- tion graph of the tokens in the document (Fig. …

Content truncated to 500 words