Paraphrase Generation via Adversarial Penalizations

Gerson Vizcarra, Jose Ochoa-Luna


Abstract
Paraphrase generation is an important problem in Natural Language Processing that has been addressed with neural network-based approaches recently. This paper presents an adversarial framework to address the paraphrase generation problem in English. Unlike previous methods, we employ the discriminator output as penalization instead of using policy gradients, and we propose a global discriminator to avoid the Monte-Carlo search. In addition, this work use and compare different settings of input representation. We compare our methods to some baselines in the Quora question pairs dataset. The results show that our framework is competitive against the previous benchmarks.
Anthology ID:
2020.wnut-1.32
Volume:
Proceedings of the Sixth Workshop on Noisy User-generated Text (W-NUT 2020)
Month:
November
Year:
2020
Address:
Online
Venue:
WNUT
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
249–259
Language:
URL:
https://aclanthology.org/2020.wnut-1.32
DOI:
10.18653/v1/2020.wnut-1.32
Bibkey:
Cite (ACL):
Gerson Vizcarra and Jose Ochoa-Luna. 2020. Paraphrase Generation via Adversarial Penalizations. In Proceedings of the Sixth Workshop on Noisy User-generated Text (W-NUT 2020), pages 249–259, Online. Association for Computational Linguistics.
Cite (Informal):
Paraphrase Generation via Adversarial Penalizations (Vizcarra & Ochoa-Luna, WNUT 2020)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-script-update/2020.wnut-1.32.pdf