Learning to Selectively Learn for Weakly-supervised Paraphrase Generation

Kaize Ding, Dingcheng Li, Alexander Hanbo Li, Xing Fan, Chenlei Guo, Yang Liu, Huan Liu


Abstract
Paraphrase generation is a longstanding NLP task that has diverse applications on downstream NLP tasks. However, the effectiveness of existing efforts predominantly relies on large amounts of golden labeled data. Though unsupervised endeavors have been proposed to alleviate this issue, they may fail to generate meaningful paraphrases due to the lack of supervision signals. In this work, we go beyond the existing paradigms and propose a novel approach to generate high-quality paraphrases with data of weak supervision. Specifically, we tackle the weakly-supervised paraphrase generation problem by: (1) obtaining abundant weakly-labeled parallel sentences via retrieval-based pseudo paraphrase expansion; and (2) developing a meta-learning framework to progressively select valuable samples for fine-tuning a pre-trained language model BART on the sentential paraphrasing task. We demonstrate that our approach achieves significant improvements over existing unsupervised approaches, and is even comparable in performance with supervised state-of-the-arts.
Anthology ID:
2021.emnlp-main.480
Volume:
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2021
Address:
Online and Punta Cana, Dominican Republic
Editors:
Marie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-tau Yih
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
5930–5940
Language:
URL:
https://aclanthology.org/2021.emnlp-main.480
DOI:
10.18653/v1/2021.emnlp-main.480
Bibkey:
Cite (ACL):
Kaize Ding, Dingcheng Li, Alexander Hanbo Li, Xing Fan, Chenlei Guo, Yang Liu, and Huan Liu. 2021. Learning to Selectively Learn for Weakly-supervised Paraphrase Generation. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 5930–5940, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
Learning to Selectively Learn for Weakly-supervised Paraphrase Generation (Ding et al., EMNLP 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.emnlp-main.480.pdf
Video:
 https://aclanthology.org/2021.emnlp-main.480.mp4
Data
MS COCO