Lexicon-enhanced transformer with spatial-aware integration for Chinese named entity recognition

Abstract Chinese Named Entity Recognition (CNER) is a fundamental and crucial task in information extraction. In recent years, pre-trained language and lexicon-based models have proven more powerful than the previous character-based models in CNER tasks. However, existing lexicon-enhanced BERT model...

Full description

Saved in:
Bibliographic Details
Main Authors: Jiachen Huang, Shuo Liu
Format: Article
Language:English
Published: Springer 2025-07-01
Series:Complex & Intelligent Systems
Subjects:
Online Access:https://doi.org/10.1007/s40747-025-01953-2
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849235525528977408
author Jiachen Huang
Shuo Liu
author_facet Jiachen Huang
Shuo Liu
author_sort Jiachen Huang
collection DOAJ
description Abstract Chinese Named Entity Recognition (CNER) is a fundamental and crucial task in information extraction. In recent years, pre-trained language and lexicon-based models have proven more powerful than the previous character-based models in CNER tasks. However, existing lexicon-enhanced BERT models neither integrate lexical knowledge into the fundamental layers of the bidirectional transformer model nor explicitly align character features with lexicon features. In this paper, we propose a spatial-aware lexicon adapter (SALA), a neural adapter capable of dynamically integrating character and lexical representations through spatial-aware attention. SALA is incorporated between the layers of BERT to inject lexical information into the deep contextual representations of corresponding character sequences. The resulting fused vectors are further trained in SALA-BERT to enhance CNER. We evaluate SALA-BERT on various Chinese NER tasks. Compared to previous state-of-the-art models, it achieves comparable or better performance.
format Article
id doaj-art-328ff9d32d5b4e9cb3c31f99f0dd4c33
institution Kabale University
issn 2199-4536
2198-6053
language English
publishDate 2025-07-01
publisher Springer
record_format Article
series Complex & Intelligent Systems
spelling doaj-art-328ff9d32d5b4e9cb3c31f99f0dd4c332025-08-20T04:02:45ZengSpringerComplex & Intelligent Systems2199-45362198-60532025-07-0111811710.1007/s40747-025-01953-2Lexicon-enhanced transformer with spatial-aware integration for Chinese named entity recognitionJiachen Huang0Shuo Liu1Aerospace Information Research Institute, Chinese Academy of ScienceAerospace Information Research Institute, Chinese Academy of ScienceAbstract Chinese Named Entity Recognition (CNER) is a fundamental and crucial task in information extraction. In recent years, pre-trained language and lexicon-based models have proven more powerful than the previous character-based models in CNER tasks. However, existing lexicon-enhanced BERT models neither integrate lexical knowledge into the fundamental layers of the bidirectional transformer model nor explicitly align character features with lexicon features. In this paper, we propose a spatial-aware lexicon adapter (SALA), a neural adapter capable of dynamically integrating character and lexical representations through spatial-aware attention. SALA is incorporated between the layers of BERT to inject lexical information into the deep contextual representations of corresponding character sequences. The resulting fused vectors are further trained in SALA-BERT to enhance CNER. We evaluate SALA-BERT on various Chinese NER tasks. Compared to previous state-of-the-art models, it achieves comparable or better performance.https://doi.org/10.1007/s40747-025-01953-2Named Entity RecognitionBERTNatural language generationInformation extraction
spellingShingle Jiachen Huang
Shuo Liu
Lexicon-enhanced transformer with spatial-aware integration for Chinese named entity recognition
Complex & Intelligent Systems
Named Entity Recognition
BERT
Natural language generation
Information extraction
title Lexicon-enhanced transformer with spatial-aware integration for Chinese named entity recognition
title_full Lexicon-enhanced transformer with spatial-aware integration for Chinese named entity recognition
title_fullStr Lexicon-enhanced transformer with spatial-aware integration for Chinese named entity recognition
title_full_unstemmed Lexicon-enhanced transformer with spatial-aware integration for Chinese named entity recognition
title_short Lexicon-enhanced transformer with spatial-aware integration for Chinese named entity recognition
title_sort lexicon enhanced transformer with spatial aware integration for chinese named entity recognition
topic Named Entity Recognition
BERT
Natural language generation
Information extraction
url https://doi.org/10.1007/s40747-025-01953-2
work_keys_str_mv AT jiachenhuang lexiconenhancedtransformerwithspatialawareintegrationforchinesenamedentityrecognition
AT shuoliu lexiconenhancedtransformerwithspatialawareintegrationforchinesenamedentityrecognition