https://scholars.lib.ntu.edu.tw/handle/123456789/489773
標題: | Multiple Text Style Transfer by using Word-level Conditional Generative Adversarial Network with Two-Phase Training. | 作者: | Lai, Chih-Te Hong, Yi-Te Chen, Hong-You Lu, Chi-Jen Lin, Shou-De SHOU-DE LIN |
公開日期: | 2019 | 起(迄)頁: | 3579-3584 | 來源出版物: | EMNLP-IJCNLP 2019 - 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint Conference on Natural Language Processing, Proceedings of the Conference | 摘要: | The objective of non-parallel text style transfer is to alter specific attributes (e.g. sentiment, mood, tense, politeness, etc) of a given text while preserving unrelated content. Adversarial training is a popular method to ensure the transferred sentences have the desired target styles. However, previous works often suffer from content leaking problem. In this paper, we propose a new adversarial training model with a word-level conditional architecture and a two-phase training procedure. By using a style-related condition architecture before generating a word, our model is able to maintain style-unrelated words while changing the others. By separating the training procedure into reconstruction and transfer phases, our model is able to balance the reconstruction and adversarial losses. We test our model on polarity sentiment transfer and multiple-attribute transfer tasks. The empirical results show that our model achieves comparable evaluation scores in both transfer accuracy and fluency but significantly outperforms other state-of-the-art models in content compatibility on three real-world datasets. © 2019 Association for Computational Linguistics |
URI: | https://scholars.lib.ntu.edu.tw/handle/123456789/489773 https://www.scopus.com/inward/record.uri?eid=2-s2.0-85084295185&partnerID=40&md5=4945c5838f80a0aafb03aad4ea4d4949 |
DOI: | 10.18653/v1/D19-1366 | SDG/關鍵字: | Network architecture; Adversarial networks; Multiple attributes; Parallel text; Real-world datasets; State of the art; Training model; Training procedures; Word level; Natural language processing systems |
顯示於: | 資訊工程學系 |
在 IR 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。