ChunkUIE: Chunked instruction-based unified information extraction.

Large language models (LLMs) have demonstrated remarkable performance across various linguistic tasks. However, existing LLMs perform inadequately in information extraction tasks for both Chinese and English. Numerous studies attempt to enhance model performance by increasing the scale of training d...

Full description

Saved in:
Bibliographic Details
Main Authors: Wei Li, Yingzhen Liu, Yinling Yang, Ting Zhang, Wei Men
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2025-01-01
Series:PLoS ONE
Online Access:https://doi.org/10.1371/journal.pone.0326764
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849319493397905408
author Wei Li
Yingzhen Liu
Yinling Yang
Ting Zhang
Wei Men
author_facet Wei Li
Yingzhen Liu
Yinling Yang
Ting Zhang
Wei Men
author_sort Wei Li
collection DOAJ
description Large language models (LLMs) have demonstrated remarkable performance across various linguistic tasks. However, existing LLMs perform inadequately in information extraction tasks for both Chinese and English. Numerous studies attempt to enhance model performance by increasing the scale of training data. However, discrepancies in the number and type of schemas used during training and evaluation can harm model effectiveness. To tackle this challenge, we propose ChunkUIE, a unified information extraction model that supports Chinese and English. We design a chunked instruction construction strategy that randomly and reproducibly divides all schemas into chunks containing an identical number of schemas. This approach ensures that the union of schemas across all chunks encompasses all schemas. By limiting the number of schemas in each instruction, this strategy effectively addresses the performance degradation caused by inconsistencies in schema counts between training and evaluation. Additionally, we construct some challenging negative schemas using a predefined hard schema dictionary, which mitigates the model's semantic confusion regarding similar schemas. Experimental results demonstrate that ChunkUIE enhances zero-shot performance in information extraction.
format Article
id doaj-art-54be924bf664493389aafa53c01e970b
institution Kabale University
issn 1932-6203
language English
publishDate 2025-01-01
publisher Public Library of Science (PLoS)
record_format Article
series PLoS ONE
spelling doaj-art-54be924bf664493389aafa53c01e970b2025-08-20T03:50:26ZengPublic Library of Science (PLoS)PLoS ONE1932-62032025-01-01206e032676410.1371/journal.pone.0326764ChunkUIE: Chunked instruction-based unified information extraction.Wei LiYingzhen LiuYinling YangTing ZhangWei MenLarge language models (LLMs) have demonstrated remarkable performance across various linguistic tasks. However, existing LLMs perform inadequately in information extraction tasks for both Chinese and English. Numerous studies attempt to enhance model performance by increasing the scale of training data. However, discrepancies in the number and type of schemas used during training and evaluation can harm model effectiveness. To tackle this challenge, we propose ChunkUIE, a unified information extraction model that supports Chinese and English. We design a chunked instruction construction strategy that randomly and reproducibly divides all schemas into chunks containing an identical number of schemas. This approach ensures that the union of schemas across all chunks encompasses all schemas. By limiting the number of schemas in each instruction, this strategy effectively addresses the performance degradation caused by inconsistencies in schema counts between training and evaluation. Additionally, we construct some challenging negative schemas using a predefined hard schema dictionary, which mitigates the model's semantic confusion regarding similar schemas. Experimental results demonstrate that ChunkUIE enhances zero-shot performance in information extraction.https://doi.org/10.1371/journal.pone.0326764
spellingShingle Wei Li
Yingzhen Liu
Yinling Yang
Ting Zhang
Wei Men
ChunkUIE: Chunked instruction-based unified information extraction.
PLoS ONE
title ChunkUIE: Chunked instruction-based unified information extraction.
title_full ChunkUIE: Chunked instruction-based unified information extraction.
title_fullStr ChunkUIE: Chunked instruction-based unified information extraction.
title_full_unstemmed ChunkUIE: Chunked instruction-based unified information extraction.
title_short ChunkUIE: Chunked instruction-based unified information extraction.
title_sort chunkuie chunked instruction based unified information extraction
url https://doi.org/10.1371/journal.pone.0326764
work_keys_str_mv AT weili chunkuiechunkedinstructionbasedunifiedinformationextraction
AT yingzhenliu chunkuiechunkedinstructionbasedunifiedinformationextraction
AT yinlingyang chunkuiechunkedinstructionbasedunifiedinformationextraction
AT tingzhang chunkuiechunkedinstructionbasedunifiedinformationextraction
AT weimen chunkuiechunkedinstructionbasedunifiedinformationextraction