Saved in:
Bibliographic Details
Main Authors: Dong, Yanfei, Deng, Lambert, Zhang, Jiazheng, Yu, Xiaodong, Lin, Ting, Gelli, Francesco, Poria, Soujanya, Lee, Wee Sun
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2405.06701
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917662841372672
author Dong, Yanfei
Deng, Lambert
Zhang, Jiazheng
Yu, Xiaodong
Lin, Ting
Gelli, Francesco
Poria, Soujanya
Lee, Wee Sun
author_facet Dong, Yanfei
Deng, Lambert
Zhang, Jiazheng
Yu, Xiaodong
Lin, Ting
Gelli, Francesco
Poria, Soujanya
Lee, Wee Sun
contents Documents that consist of diverse templates and exhibit complex spatial structures pose a challenge for document entity classification. We propose KNN-former, which incorporates a new kind of spatial bias in attention calculation based on the K-nearest-neighbor (KNN) graph of document entities. We limit entities' attention only to their local radius defined by the KNN graph. We also use combinatorial matching to address the one-to-one mapping property that exists in many documents, where one field has only one corresponding entity. Moreover, our method is highly parameter-efficient compared to existing approaches in terms of the number of trainable parameters. Despite this, experiments across various datasets show our method outperforms baselines in most entity types. Many real-world documents exhibit combinatorial properties which can be leveraged as inductive biases to improve extraction accuracy, but existing datasets do not cover these documents. To facilitate future research into these types of documents, we release a new ID document dataset that covers diverse templates and languages. We also release enhanced annotations for an existing dataset.
format Preprint
id arxiv_https___arxiv_org_abs_2405_06701
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Lightweight Spatial Modeling for Combinatorial Information Extraction From Documents
Dong, Yanfei
Deng, Lambert
Zhang, Jiazheng
Yu, Xiaodong
Lin, Ting
Gelli, Francesco
Poria, Soujanya
Lee, Wee Sun
Computation and Language
Artificial Intelligence
Documents that consist of diverse templates and exhibit complex spatial structures pose a challenge for document entity classification. We propose KNN-former, which incorporates a new kind of spatial bias in attention calculation based on the K-nearest-neighbor (KNN) graph of document entities. We limit entities' attention only to their local radius defined by the KNN graph. We also use combinatorial matching to address the one-to-one mapping property that exists in many documents, where one field has only one corresponding entity. Moreover, our method is highly parameter-efficient compared to existing approaches in terms of the number of trainable parameters. Despite this, experiments across various datasets show our method outperforms baselines in most entity types. Many real-world documents exhibit combinatorial properties which can be leveraged as inductive biases to improve extraction accuracy, but existing datasets do not cover these documents. To facilitate future research into these types of documents, we release a new ID document dataset that covers diverse templates and languages. We also release enhanced annotations for an existing dataset.
title Lightweight Spatial Modeling for Combinatorial Information Extraction From Documents
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2405.06701