Advancing Few-Shot Named Entity Recognition with Large Language Model

Few-shot named entity recognition (NER) involves identifying specific entities using limited data. Metric learning-based methods, which compute token-level similarities between query and support sets to identify target entities, have demonstrated remarkable performance in this task. However, their e...

Full description

Saved in:
Bibliographic Details
Main Authors: Yuhui Xiao, Jianjian Zou, Qun Yang
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/7/3838
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850188257903509504
author Yuhui Xiao
Jianjian Zou
Qun Yang
author_facet Yuhui Xiao
Jianjian Zou
Qun Yang
author_sort Yuhui Xiao
collection DOAJ
description Few-shot named entity recognition (NER) involves identifying specific entities using limited data. Metric learning-based methods, which compute token-level similarities between query and support sets to identify target entities, have demonstrated remarkable performance in this task. However, their effectiveness deteriorates when the distribution of the support set differs from that of the query set. To address this issue, we propose a novel approach that leverages the synergy between the large language model (LLM) and the metric learning-based few-shot NER approach. Specifically, we use the LLM to refine low-confidence predictions produced by the metric learning-based few-shot NER model, thus improving overall recognition accuracy. To further reduce the difficulty of entity classification, we introduce multiple label-filtering strategies to reduce the difficulty for LLMs in performing entity classification. Furthermore, we explore the impact of prompt design on enhancing NER performance. Experimental results show that the proposed method increases the micro-F1 score on Few-NERD and CrossNER by 0.86% and 4.9%, respectively, compared to previous state-of-the-art methods.
format Article
id doaj-art-937072bfb5244ecfa6034581a7eaedfb
institution OA Journals
issn 2076-3417
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-937072bfb5244ecfa6034581a7eaedfb2025-08-20T02:15:55ZengMDPI AGApplied Sciences2076-34172025-04-01157383810.3390/app15073838Advancing Few-Shot Named Entity Recognition with Large Language ModelYuhui Xiao0Jianjian Zou1Qun Yang2College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, ChinaCollege of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, ChinaCollege of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, ChinaFew-shot named entity recognition (NER) involves identifying specific entities using limited data. Metric learning-based methods, which compute token-level similarities between query and support sets to identify target entities, have demonstrated remarkable performance in this task. However, their effectiveness deteriorates when the distribution of the support set differs from that of the query set. To address this issue, we propose a novel approach that leverages the synergy between the large language model (LLM) and the metric learning-based few-shot NER approach. Specifically, we use the LLM to refine low-confidence predictions produced by the metric learning-based few-shot NER model, thus improving overall recognition accuracy. To further reduce the difficulty of entity classification, we introduce multiple label-filtering strategies to reduce the difficulty for LLMs in performing entity classification. Furthermore, we explore the impact of prompt design on enhancing NER performance. Experimental results show that the proposed method increases the micro-F1 score on Few-NERD and CrossNER by 0.86% and 4.9%, respectively, compared to previous state-of-the-art methods.https://www.mdpi.com/2076-3417/15/7/3838few-shot named entity recognitionlarge language modelsprompt engineering
spellingShingle Yuhui Xiao
Jianjian Zou
Qun Yang
Advancing Few-Shot Named Entity Recognition with Large Language Model
Applied Sciences
few-shot named entity recognition
large language models
prompt engineering
title Advancing Few-Shot Named Entity Recognition with Large Language Model
title_full Advancing Few-Shot Named Entity Recognition with Large Language Model
title_fullStr Advancing Few-Shot Named Entity Recognition with Large Language Model
title_full_unstemmed Advancing Few-Shot Named Entity Recognition with Large Language Model
title_short Advancing Few-Shot Named Entity Recognition with Large Language Model
title_sort advancing few shot named entity recognition with large language model
topic few-shot named entity recognition
large language models
prompt engineering
url https://www.mdpi.com/2076-3417/15/7/3838
work_keys_str_mv AT yuhuixiao advancingfewshotnamedentityrecognitionwithlargelanguagemodel
AT jianjianzou advancingfewshotnamedentityrecognitionwithlargelanguagemodel
AT qunyang advancingfewshotnamedentityrecognitionwithlargelanguagemodel