Abstract
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%.
Highlights
Style transfer plays important role in many subfields of artificial intelligence (AI), such as natural language processing (NLP) [1]–[3] and computer vision (CV) [4]–[8], as it reflects the ability of intelligent systems to generate novel contents
To address the above issues, we propose a model called unified generative adversarial networks(UGAN) for multidirectional text style transfer
We propose Unified Generative Adversarial Networks (UGAN), a novel generation adversarial networks that uses a single generator and discriminator to learn the mapping between multiple attributes and transfer style effectively among all attributes
Summary
Style transfer plays important role in many subfields of artificial intelligence (AI), such as natural language processing (NLP) [1]–[3] and computer vision (CV) [4]–[8], as it reflects the ability of intelligent systems to generate novel contents. Text style transfer is an important branch of style transfer, it aims to rephrase the input text into the desired style while preserving its content. Deep neural networks have become the mainstream method of text style transfer. Many previous work [9]–[12] use encoder-decoder framework to implement style transfer. The encoder maps the text into a style-independent latent representation, and the decoder generates a new text with the same content but a different style from the disentangled latent representation plus a style variable. Due to lacking paired sentence, an adversarial loss [13] is used in the latent space to discourage encoding style in the latent
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.