An end-to-end generative architecture for paraphrase generation

Qian Yang, Zhouyuan Huo, Dinghan Shen, Yong Cheng, Wenlin Wang, Guoyin Wang, Lawrence Carin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

8 Scopus citations

Abstract

Generating high-quality paraphrases is a fundamental yet challenging natural language processing task. Despite the effectiveness of previous work based on generative models, there remain problems with exposure bias in recurrent neural networks, and often a failure to generate realistic sentences. To overcome these challenges, we propose the first end-to-end conditional generative architecture for generating paraphrases via adversarial training, which does not depend on extra linguistic information. Extensive experiments on four public datasets demonstrate the proposed method achieves state-of-the-art results, outperforming previous generative architectures on both automatic metrics (BLEU, METEOR, and TER) and human evaluations.
Original languageEnglish (US)
Title of host publicationEMNLP-IJCNLP 2019 - 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint Conference on Natural Language Processing, Proceedings of the Conference
PublisherAssociation for Computational Linguistics
Pages3132-3142
Number of pages11
ISBN (Print)9781950737901
StatePublished - Jan 1 2020
Externally publishedYes

Fingerprint

Dive into the research topics of 'An end-to-end generative architecture for paraphrase generation'. Together they form a unique fingerprint.

Cite this