KeyEE: Enhancing Low-Resource Generative Event Extraction with Auxiliary Keyword Sub-Prompt
Event Extraction (EE) is a key task in information extraction, which requires high-quality annotated data that are often costly to obtain. Traditional classification-based methods suffer from low-resource scenarios due to the lack of label semantics and fine-grained annotations. While recent approac...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Tsinghua University Press
2024-06-01
|
| Series: | Big Data Mining and Analytics |
| Subjects: | |
| Online Access: | https://www.sciopen.com/article/10.26599/BDMA.2023.9020036 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832569354359668736 |
|---|---|
| author | Junwen Duan Xincheng Liao Ying An Jianxin Wang |
| author_facet | Junwen Duan Xincheng Liao Ying An Jianxin Wang |
| author_sort | Junwen Duan |
| collection | DOAJ |
| description | Event Extraction (EE) is a key task in information extraction, which requires high-quality annotated data that are often costly to obtain. Traditional classification-based methods suffer from low-resource scenarios due to the lack of label semantics and fine-grained annotations. While recent approaches have endeavored to address EE through a more data-efficient generative process, they often overlook event keywords, which are vital for EE. To tackle these challenges, we introduce KeyEE, a multi-prompt learning strategy that improves low-resource event extraction by Event Keywords Extraction(EKE). We suggest employing an auxiliary EKE sub-prompt and concurrently training both EE and EKE with a shared pre-trained language model. With the auxiliary sub-prompt, KeyEE learns event keywords knowledge implicitly, thereby reducing the dependence on annotated data. Furthermore, we investigate and analyze various EKE sub-prompt strategies to encourage further research in this area. Our experiments on benchmark datasets ACE2005 and ERE show that KeyEE achieves significant improvement in low-resource settings and sets new state-of-the-art results. |
| format | Article |
| id | doaj-art-bdbb91c1c50043ccb22e91407bd5e58f |
| institution | Kabale University |
| issn | 2096-0654 |
| language | English |
| publishDate | 2024-06-01 |
| publisher | Tsinghua University Press |
| record_format | Article |
| series | Big Data Mining and Analytics |
| spelling | doaj-art-bdbb91c1c50043ccb22e91407bd5e58f2025-02-02T22:18:05ZengTsinghua University PressBig Data Mining and Analytics2096-06542024-06-017254756010.26599/BDMA.2023.9020036KeyEE: Enhancing Low-Resource Generative Event Extraction with Auxiliary Keyword Sub-PromptJunwen Duan0Xincheng Liao1Ying An2Jianxin Wang3Hunan Key Laboratory of Bioinformatics, School of Computer Science and Engineering, Central South University, Changsha 410083, ChinaHunan Key Laboratory of Bioinformatics, School of Computer Science and Engineering, Central South University, Changsha 410083, ChinaBig Data Institute, Central South University, Changsha 410083, ChinaHunan Key Laboratory of Bioinformatics, School of Computer Science and Engineering, Central South University, Changsha 410083, ChinaEvent Extraction (EE) is a key task in information extraction, which requires high-quality annotated data that are often costly to obtain. Traditional classification-based methods suffer from low-resource scenarios due to the lack of label semantics and fine-grained annotations. While recent approaches have endeavored to address EE through a more data-efficient generative process, they often overlook event keywords, which are vital for EE. To tackle these challenges, we introduce KeyEE, a multi-prompt learning strategy that improves low-resource event extraction by Event Keywords Extraction(EKE). We suggest employing an auxiliary EKE sub-prompt and concurrently training both EE and EKE with a shared pre-trained language model. With the auxiliary sub-prompt, KeyEE learns event keywords knowledge implicitly, thereby reducing the dependence on annotated data. Furthermore, we investigate and analyze various EKE sub-prompt strategies to encourage further research in this area. Our experiments on benchmark datasets ACE2005 and ERE show that KeyEE achieves significant improvement in low-resource settings and sets new state-of-the-art results.https://www.sciopen.com/article/10.26599/BDMA.2023.9020036natural language processingevent extraction (ee)multi-prompt learning (mpl)low-resource |
| spellingShingle | Junwen Duan Xincheng Liao Ying An Jianxin Wang KeyEE: Enhancing Low-Resource Generative Event Extraction with Auxiliary Keyword Sub-Prompt Big Data Mining and Analytics natural language processing event extraction (ee) multi-prompt learning (mpl) low-resource |
| title | KeyEE: Enhancing Low-Resource Generative Event Extraction with Auxiliary Keyword Sub-Prompt |
| title_full | KeyEE: Enhancing Low-Resource Generative Event Extraction with Auxiliary Keyword Sub-Prompt |
| title_fullStr | KeyEE: Enhancing Low-Resource Generative Event Extraction with Auxiliary Keyword Sub-Prompt |
| title_full_unstemmed | KeyEE: Enhancing Low-Resource Generative Event Extraction with Auxiliary Keyword Sub-Prompt |
| title_short | KeyEE: Enhancing Low-Resource Generative Event Extraction with Auxiliary Keyword Sub-Prompt |
| title_sort | keyee enhancing low resource generative event extraction with auxiliary keyword sub prompt |
| topic | natural language processing event extraction (ee) multi-prompt learning (mpl) low-resource |
| url | https://www.sciopen.com/article/10.26599/BDMA.2023.9020036 |
| work_keys_str_mv | AT junwenduan keyeeenhancinglowresourcegenerativeeventextractionwithauxiliarykeywordsubprompt AT xinchengliao keyeeenhancinglowresourcegenerativeeventextractionwithauxiliarykeywordsubprompt AT yingan keyeeenhancinglowresourcegenerativeeventextractionwithauxiliarykeywordsubprompt AT jianxinwang keyeeenhancinglowresourcegenerativeeventextractionwithauxiliarykeywordsubprompt |