Principled Paraphrase Generation with Parallel Corpora

Aitor Ormazabal, Mikel Artetxe, Aitor Soroa, Gorka Labaka, Eneko Agirre


Abstract
Round-trip Machine Translation (MT) is a popular choice for paraphrase generation, which leverages readily available parallel corpora for supervision. In this paper, we formalize the implicit similarity function induced by this approach, and show that it is susceptible to non-paraphrase pairs sharing a single ambiguous translation. Based on these insights, we design an alternative similarity metric that mitigates this issue by requiring the entire translation distribution to match, and implement a relaxation of it through the Information Bottleneck method. Our approach incorporates an adversarial term into MT training in order to learn representations that encode as much information about the reference translation as possible, while keeping as little information about the input as possible. Paraphrases can be generated by decoding back to the source from this representation, without having to generate pivot translations. In addition to being more principled and efficient than round-trip MT, our approach offers an adjustable parameter to control the fidelity-diversity trade-off, and obtains better results in our experiments.
Anthology ID:
2022.acl-long.114
Volume:
Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
May
Year:
2022
Address:
Dublin, Ireland
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1621–1638
Language:
URL:
https://aclanthology.org/2022.acl-long.114
DOI:
10.18653/v1/2022.acl-long.114
Bibkey:
Cite (ACL):
Aitor Ormazabal, Mikel Artetxe, Aitor Soroa, Gorka Labaka, and Eneko Agirre. 2022. Principled Paraphrase Generation with Parallel Corpora. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1621–1638, Dublin, Ireland. Association for Computational Linguistics.
Cite (Informal):
Principled Paraphrase Generation with Parallel Corpora (Ormazabal et al., ACL 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.acl-long.114.pdf
Code
 aitorormazabal/paraphrasing-from-parallel