SE-Prompt: Exploring Semantic Enhancement with Prompt Tuning for Relation Extraction

Cai Wang; Dongyang Li; Xiaofeng He

doi:10.1007/978-3-031-46674-8_8

SE-Prompt: Exploring Semantic Enhancement with Prompt Tuning for Relation Extraction

Cai Wang
, Dongyang Li
, Xiaofeng He^*

^*Corresponding author for this work

School of Computer Science and Technology

East China Normal University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

Compared to traditional supervised learning methods, utilizing prompt tuning for relation extraction tasks is a challenging endeavor in the real world. By inserting a template segment into the input, prompt tuning has proven effective for certain classification tasks. However, applying prompt tuning to relation extraction tasks, which involve mapping multiple words to a single label, poses challenges due to difficulties in precisely defining a template and mapping labels to the appropriate words. Prior approaches do not take full advantage of entities and have also overlooked the semantic connections between words in relation label. To address these limitations, we propose a semantic enhancement with prompt (SE-Prompt) which integrates entity and relation knowledge by incorporating two main contributions: semantic enhancement and subject-object relation refinement. These methods empower our model to effectively leverage relation labels and tap into the knowledge contained in pre-trained models. Our experiments on three datasets, under both fully supervised and low-resource settings demonstrate the effectiveness of our approach for relation extraction.

Original language	English
Title of host publication	Advanced Data Mining and Applications - 19th International Conference, ADMA 2023, Proceedings
Editors	Xiaochun Yang, Bin Wang, Heru Suhartanto, Guoren Wang, Jing Jiang, Bing Li, Huaijie Zhu, Ningning Cui
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	109-122
Number of pages	14
ISBN (Print)	9783031466731
DOIs	https://doi.org/10.1007/978-3-031-46674-8_8
State	Published - 2023
Event	19th International Conference on Advanced Data Mining and Applications, ADMA 2023 - Shenyang, China Duration: 21 Aug 2023 → 23 Aug 2023

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	14179 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	19th International Conference on Advanced Data Mining and Applications, ADMA 2023
Country/Territory	China
City	Shenyang
Period	21/08/23 → 23/08/23

Keywords

Entity Expansion
Prompt Tuning
Relation Extraction
Semantic Enhancement

Access to Document

10.1007/978-3-031-46674-8_8

Cite this

Wang, C., Li, D., & He, X. (2023). SE-Prompt: Exploring Semantic Enhancement with Prompt Tuning for Relation Extraction. In X. Yang, B. Wang, H. Suhartanto, G. Wang, J. Jiang, B. Li, H. Zhu, & N. Cui (Eds.), Advanced Data Mining and Applications - 19th International Conference, ADMA 2023, Proceedings (pp. 109-122). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 14179 LNAI). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-46674-8_8

Wang, Cai ; Li, Dongyang ; He, Xiaofeng. / SE-Prompt : Exploring Semantic Enhancement with Prompt Tuning for Relation Extraction. Advanced Data Mining and Applications - 19th International Conference, ADMA 2023, Proceedings. editor / Xiaochun Yang ; Bin Wang ; Heru Suhartanto ; Guoren Wang ; Jing Jiang ; Bing Li ; Huaijie Zhu ; Ningning Cui. Springer Science and Business Media Deutschland GmbH, 2023. pp. 109-122 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{d0981fca700949769d77830c93c5bd99,

title = "SE-Prompt: Exploring Semantic Enhancement with Prompt Tuning for Relation Extraction",

abstract = "Compared to traditional supervised learning methods, utilizing prompt tuning for relation extraction tasks is a challenging endeavor in the real world. By inserting a template segment into the input, prompt tuning has proven effective for certain classification tasks. However, applying prompt tuning to relation extraction tasks, which involve mapping multiple words to a single label, poses challenges due to difficulties in precisely defining a template and mapping labels to the appropriate words. Prior approaches do not take full advantage of entities and have also overlooked the semantic connections between words in relation label. To address these limitations, we propose a semantic enhancement with prompt (SE-Prompt) which integrates entity and relation knowledge by incorporating two main contributions: semantic enhancement and subject-object relation refinement. These methods empower our model to effectively leverage relation labels and tap into the knowledge contained in pre-trained models. Our experiments on three datasets, under both fully supervised and low-resource settings demonstrate the effectiveness of our approach for relation extraction.",

keywords = "Entity Expansion, Prompt Tuning, Relation Extraction, Semantic Enhancement",

author = "Cai Wang and Dongyang Li and Xiaofeng He",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.; 19th International Conference on Advanced Data Mining and Applications, ADMA 2023 ; Conference date: 21-08-2023 Through 23-08-2023",

year = "2023",

doi = "10.1007/978-3-031-46674-8\_8",

language = "英语",

isbn = "9783031466731",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "109--122",

editor = "Xiaochun Yang and Bin Wang and Heru Suhartanto and Guoren Wang and Jing Jiang and Bing Li and Huaijie Zhu and Ningning Cui",

booktitle = "Advanced Data Mining and Applications - 19th International Conference, ADMA 2023, Proceedings",

address = "德国",

}

Wang, C, Li, D & He, X 2023, SE-Prompt: Exploring Semantic Enhancement with Prompt Tuning for Relation Extraction. in X Yang, B Wang, H Suhartanto, G Wang, J Jiang, B Li, H Zhu & N Cui (eds), Advanced Data Mining and Applications - 19th International Conference, ADMA 2023, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 14179 LNAI, Springer Science and Business Media Deutschland GmbH, pp. 109-122, 19th International Conference on Advanced Data Mining and Applications, ADMA 2023, Shenyang, China, 21/08/23. https://doi.org/10.1007/978-3-031-46674-8_8

SE-Prompt: Exploring Semantic Enhancement with Prompt Tuning for Relation Extraction. / Wang, Cai; Li, Dongyang; He, Xiaofeng.
Advanced Data Mining and Applications - 19th International Conference, ADMA 2023, Proceedings. ed. / Xiaochun Yang; Bin Wang; Heru Suhartanto; Guoren Wang; Jing Jiang; Bing Li; Huaijie Zhu; Ningning Cui. Springer Science and Business Media Deutschland GmbH, 2023. p. 109-122 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 14179 LNAI).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - SE-Prompt

T2 - 19th International Conference on Advanced Data Mining and Applications, ADMA 2023

AU - Wang, Cai

AU - Li, Dongyang

AU - He, Xiaofeng

N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.

PY - 2023

Y1 - 2023

N2 - Compared to traditional supervised learning methods, utilizing prompt tuning for relation extraction tasks is a challenging endeavor in the real world. By inserting a template segment into the input, prompt tuning has proven effective for certain classification tasks. However, applying prompt tuning to relation extraction tasks, which involve mapping multiple words to a single label, poses challenges due to difficulties in precisely defining a template and mapping labels to the appropriate words. Prior approaches do not take full advantage of entities and have also overlooked the semantic connections between words in relation label. To address these limitations, we propose a semantic enhancement with prompt (SE-Prompt) which integrates entity and relation knowledge by incorporating two main contributions: semantic enhancement and subject-object relation refinement. These methods empower our model to effectively leverage relation labels and tap into the knowledge contained in pre-trained models. Our experiments on three datasets, under both fully supervised and low-resource settings demonstrate the effectiveness of our approach for relation extraction.

AB - Compared to traditional supervised learning methods, utilizing prompt tuning for relation extraction tasks is a challenging endeavor in the real world. By inserting a template segment into the input, prompt tuning has proven effective for certain classification tasks. However, applying prompt tuning to relation extraction tasks, which involve mapping multiple words to a single label, poses challenges due to difficulties in precisely defining a template and mapping labels to the appropriate words. Prior approaches do not take full advantage of entities and have also overlooked the semantic connections between words in relation label. To address these limitations, we propose a semantic enhancement with prompt (SE-Prompt) which integrates entity and relation knowledge by incorporating two main contributions: semantic enhancement and subject-object relation refinement. These methods empower our model to effectively leverage relation labels and tap into the knowledge contained in pre-trained models. Our experiments on three datasets, under both fully supervised and low-resource settings demonstrate the effectiveness of our approach for relation extraction.

KW - Entity Expansion

KW - Prompt Tuning

KW - Relation Extraction

KW - Semantic Enhancement

UR - https://www.scopus.com/pages/publications/85177432643

U2 - 10.1007/978-3-031-46674-8_8

DO - 10.1007/978-3-031-46674-8_8

M3 - 会议稿件

AN - SCOPUS:85177432643

SN - 9783031466731

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 109

EP - 122

BT - Advanced Data Mining and Applications - 19th International Conference, ADMA 2023, Proceedings

A2 - Yang, Xiaochun

A2 - Wang, Bin

A2 - Suhartanto, Heru

A2 - Wang, Guoren

A2 - Jiang, Jing

A2 - Li, Bing

A2 - Zhu, Huaijie

A2 - Cui, Ningning

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 21 August 2023 through 23 August 2023

ER -

Wang C, Li D, He X. SE-Prompt: Exploring Semantic Enhancement with Prompt Tuning for Relation Extraction. In Yang X, Wang B, Suhartanto H, Wang G, Jiang J, Li B, Zhu H, Cui N, editors, Advanced Data Mining and Applications - 19th International Conference, ADMA 2023, Proceedings. Springer Science and Business Media Deutschland GmbH. 2023. p. 109-122. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-031-46674-8_8