Prompt Tuning Techniques for Chinese Idiom Recommendation
Chinese idioms pose significant challenges in natural language processing (NLP) due to their complex, non-compositional nature and frequent embedded cultural and historical meanings. This study investigates prompt tuning techniques for Chinese idiom recommendation, exploring multiple-choice (MC) pro...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10965689/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849712614547914752 |
|---|---|
| author | Shun-Ming Wang I-Fang Su Yu-Chi Chung |
| author_facet | Shun-Ming Wang I-Fang Su Yu-Chi Chung |
| author_sort | Shun-Ming Wang |
| collection | DOAJ |
| description | Chinese idioms pose significant challenges in natural language processing (NLP) due to their complex, non-compositional nature and frequent embedded cultural and historical meanings. This study investigates prompt tuning techniques for Chinese idiom recommendation, exploring multiple-choice (MC) prompts, binary classification (BC) prompts, and prompt ensembling strategies. We also introduce an innovative dynamic candidate sampling strategy (DCSS) technique designed to mitigate the overfitting issues commonly encountered with prompt tuning methods on Chinese idiom datasets. Our experimental results demonstrate that prompt tuning combined with ensembling methods significantly improves model performance across multiple datasets. The proposed methods outperform the state-of-the-art (SOTA) methods while maintaining training and inference efficiency. Moreover, we show that prompt tuning can effectively be generalized for other NLP tasks, such as sentiment analysis. |
| format | Article |
| id | doaj-art-4e0400a2428d4123b90eff664daef39a |
| institution | DOAJ |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-4e0400a2428d4123b90eff664daef39a2025-08-20T03:14:13ZengIEEEIEEE Access2169-35362025-01-0113680376805110.1109/ACCESS.2025.356118810965689Prompt Tuning Techniques for Chinese Idiom RecommendationShun-Ming Wang0I-Fang Su1https://orcid.org/0000-0002-6785-5906Yu-Chi Chung2https://orcid.org/0000-0001-7488-9998Department of Industrial Engineering and Management, National Kaohsiung University of Science and Technology, Kaohsiung City, Sanmin, TaiwanDepartment of Computer Science and Information Engineering, National University of Tainan, Tainan City, West Central District, TaiwanDepartment of Industrial Engineering and Management, National Kaohsiung University of Science and Technology, Kaohsiung City, Sanmin, TaiwanChinese idioms pose significant challenges in natural language processing (NLP) due to their complex, non-compositional nature and frequent embedded cultural and historical meanings. This study investigates prompt tuning techniques for Chinese idiom recommendation, exploring multiple-choice (MC) prompts, binary classification (BC) prompts, and prompt ensembling strategies. We also introduce an innovative dynamic candidate sampling strategy (DCSS) technique designed to mitigate the overfitting issues commonly encountered with prompt tuning methods on Chinese idiom datasets. Our experimental results demonstrate that prompt tuning combined with ensembling methods significantly improves model performance across multiple datasets. The proposed methods outperform the state-of-the-art (SOTA) methods while maintaining training and inference efficiency. Moreover, we show that prompt tuning can effectively be generalized for other NLP tasks, such as sentiment analysis.https://ieeexplore.ieee.org/document/10965689/Chinese idiom recommendationdeep learningprompt engineering/tuningpre-trained language modelNLP |
| spellingShingle | Shun-Ming Wang I-Fang Su Yu-Chi Chung Prompt Tuning Techniques for Chinese Idiom Recommendation IEEE Access Chinese idiom recommendation deep learning prompt engineering/tuning pre-trained language model NLP |
| title | Prompt Tuning Techniques for Chinese Idiom Recommendation |
| title_full | Prompt Tuning Techniques for Chinese Idiom Recommendation |
| title_fullStr | Prompt Tuning Techniques for Chinese Idiom Recommendation |
| title_full_unstemmed | Prompt Tuning Techniques for Chinese Idiom Recommendation |
| title_short | Prompt Tuning Techniques for Chinese Idiom Recommendation |
| title_sort | prompt tuning techniques for chinese idiom recommendation |
| topic | Chinese idiom recommendation deep learning prompt engineering/tuning pre-trained language model NLP |
| url | https://ieeexplore.ieee.org/document/10965689/ |
| work_keys_str_mv | AT shunmingwang prompttuningtechniquesforchineseidiomrecommendation AT ifangsu prompttuningtechniquesforchineseidiomrecommendation AT yuchichung prompttuningtechniquesforchineseidiomrecommendation |