Knowledge-Guided Prompt Learning for Few-Shot Text Classification
Recently, prompt-based learning has shown impressive performance on various natural language processing tasks in few-shot scenarios. The previous study of knowledge probing showed that the success of prompt learning contributes to the implicit knowledge stored in pre-trained language models. However...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-03-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/12/6/1486 |
| _version_ | 1797612209769021440 |
|---|---|
| author | Liangguo Wang Ruoyu Chen Li Li |
| author_facet | Liangguo Wang Ruoyu Chen Li Li |
| author_sort | Liangguo Wang |
| collection | DOAJ |
| description | Recently, prompt-based learning has shown impressive performance on various natural language processing tasks in few-shot scenarios. The previous study of knowledge probing showed that the success of prompt learning contributes to the implicit knowledge stored in pre-trained language models. However, how this implicit knowledge helps solve downstream tasks remains unclear. In this work, we propose a knowledge-guided prompt learning method that can reveal relevant knowledge for text classification. Specifically, a knowledge prompting template and two multi-task frameworks were designed, respectively. The experiments demonstrated the superiority of combining knowledge and prompt learning in few-shot text classification. |
| first_indexed | 2024-03-11T06:38:03Z |
| format | Article |
| id | doaj.art-acdcb55502bb43f996bc91b5bb7ea50a |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-11T06:38:03Z |
| publishDate | 2023-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-acdcb55502bb43f996bc91b5bb7ea50a2023-11-17T10:46:12ZengMDPI AGElectronics2079-92922023-03-01126148610.3390/electronics12061486Knowledge-Guided Prompt Learning for Few-Shot Text ClassificationLiangguo Wang0Ruoyu Chen1Li Li2School of Computer Science, Beijing Information Science & Technology University, Beijing 100101, ChinaSchool of Computer Science, Beijing Information Science & Technology University, Beijing 100101, ChinaSchool of Computer Science, Beijing Information Science & Technology University, Beijing 100101, ChinaRecently, prompt-based learning has shown impressive performance on various natural language processing tasks in few-shot scenarios. The previous study of knowledge probing showed that the success of prompt learning contributes to the implicit knowledge stored in pre-trained language models. However, how this implicit knowledge helps solve downstream tasks remains unclear. In this work, we propose a knowledge-guided prompt learning method that can reveal relevant knowledge for text classification. Specifically, a knowledge prompting template and two multi-task frameworks were designed, respectively. The experiments demonstrated the superiority of combining knowledge and prompt learning in few-shot text classification.https://www.mdpi.com/2079-9292/12/6/1486knowledge-guidedprompt learningmulti-task learningtext classification |
| spellingShingle | Liangguo Wang Ruoyu Chen Li Li Knowledge-Guided Prompt Learning for Few-Shot Text Classification Electronics knowledge-guided prompt learning multi-task learning text classification |
| title | Knowledge-Guided Prompt Learning for Few-Shot Text Classification |
| title_full | Knowledge-Guided Prompt Learning for Few-Shot Text Classification |
| title_fullStr | Knowledge-Guided Prompt Learning for Few-Shot Text Classification |
| title_full_unstemmed | Knowledge-Guided Prompt Learning for Few-Shot Text Classification |
| title_short | Knowledge-Guided Prompt Learning for Few-Shot Text Classification |
| title_sort | knowledge guided prompt learning for few shot text classification |
| topic | knowledge-guided prompt learning multi-task learning text classification |
| url | https://www.mdpi.com/2079-9292/12/6/1486 |
| work_keys_str_mv | AT liangguowang knowledgeguidedpromptlearningforfewshottextclassification AT ruoyuchen knowledgeguidedpromptlearningforfewshottextclassification AT lili knowledgeguidedpromptlearningforfewshottextclassification |