UGAN: Unified Generative Adversarial Networks for Multidirectional Text Style Transfer
| Autor: | Bo Liu, Xiaodong Wang, Wei Yu, Tao Chang, Xiaoting Guo, Yang He |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Multidirectional text style transfer
General Computer Science Computer science 02 engineering and technology Machine learning computer.software_genre Field (computer science) Style (sociolinguistics) Task (project management) 03 medical and health sciences Adversarial system 0302 clinical medicine Transfer (computing) 0202 electrical engineering electronic engineering information engineering General Materials Science Structure (mathematical logic) unified generative adversarial networks business.industry General Engineering Benchmark (computing) 020201 artificial intelligence & image processing lcsh:Electrical engineering. Electronics. Nuclear engineering Artificial intelligence generative adversarial networks business lcsh:TK1-9971 computer 030217 neurology & neurosurgery Generative grammar |
| Zdroj: | IEEE Access, Vol 8, Pp 55170-55180 (2020) |
| ISSN: | 2169-3536 |
| DOI: | 10.1109/access.2020.2980898 |
| Popis: | Recently, text style transfer has become a very hot research topic in the field of natural language processing. However, the conventional text style transfer is unidirectional, and it is not possible to obtain a model with multidirectional transformations through training once. To address this limitation, we propose a new task called multidirectional text style transfer. It aims to use a single model to transfer the underlying style of text among multiple style attributes and keep its main content unchanged. In this paper, we propose Unified Generative Adversarial Networks (UGAN), a practical approach that combines target vector and generative adversarial techniques to perform multidirectional text style transfer. Our model allows simultaneous training of multi-attribute data on a single network. Such unified structure makes our model more efficient and flexible than existing approaches. We demonstrate the superiority of our approach on three benchmark datasets. Experimental results show that our method not only outperforms other baselines, but also reduces training time by an average of 13%. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|