Semantic Matching for Sequence-to-Sequence Learning

Ruiyi Zhang, Changyou Chen, Xinyuan Zhang, Ke Bai, Lawrence Carin


Abstract
In sequence-to-sequence models, classical optimal transport (OT) can be applied to semantically match generated sentences with target sentences. However, in non-parallel settings, target sentences are usually unavailable. To tackle this issue without losing the benefits of classical OT, we present a semantic matching scheme based on the Optimal Partial Transport (OPT). Specifically, our approach partially matches semantically meaningful words between source and partial target sequences. To overcome the difficulty of detecting active regions in OPT (corresponding to the words needed to be matched), we further exploit prior knowledge to perform partial matching. Extensive experiments are conducted to evaluate the proposed approach, showing consistent improvements over sequence-to-sequence tasks.
Anthology ID:
2020.findings-emnlp.21
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2020
Month:
November
Year:
2020
Address:
Online
Editors:
Trevor Cohn, Yulan He, Yang Liu
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
212–222
Language:
URL:
https://aclanthology.org/2020.findings-emnlp.21
DOI:
10.18653/v1/2020.findings-emnlp.21
Bibkey:
Cite (ACL):
Ruiyi Zhang, Changyou Chen, Xinyuan Zhang, Ke Bai, and Lawrence Carin. 2020. Semantic Matching for Sequence-to-Sequence Learning. In Findings of the Association for Computational Linguistics: EMNLP 2020, pages 212–222, Online. Association for Computational Linguistics.
Cite (Informal):
Semantic Matching for Sequence-to-Sequence Learning (Zhang et al., Findings 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.findings-emnlp.21.pdf
Optional supplementary material:
 2020.findings-emnlp.21.OptionalSupplementaryMaterial.bbl
Data
MS COCO