Abstract
In Neural Machine Translation, using word-level tokens leads to degradation in translation quality. The dominant approaches use subword-level tokens, but this increases the length of the sequences and makes it difficult to profit from word-level information such as POS tags or semantic dependencies. We propose a modification to the Transformer model to combine subword-level representations into word-level ones in the first layers of the encoder, reducing the effective length of the sequences in the following layers and providing a natural point to incorporate extra word-level information. Our experiments show that this approach maintains the translation quality with respect to the normal Transformer model when no extra word-level information is injected and that it is superior to the currently dominant method for incorporating word-level source language information to models based on subword-level vocabularies.- Anthology ID:
- 2020.acl-srw.10
- Volume:
- Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics: Student Research Workshop
- Month:
- July
- Year:
- 2020
- Address:
- Online
- Venue:
- ACL
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 66–71
- Language:
- URL:
- https://aclanthology.org/2020.acl-srw.10
- DOI:
- 10.18653/v1/2020.acl-srw.10
- Cite (ACL):
- Noe Casas, Marta R. Costa-jussà, and José A. R. Fonollosa. 2020. Combining Subword Representations into Word-level Representations in the Transformer Architecture. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics: Student Research Workshop, pages 66–71, Online. Association for Computational Linguistics.
- Cite (Informal):
- Combining Subword Representations into Word-level Representations in the Transformer Architecture (Casas et al., ACL 2020)
- PDF:
- https://preview.aclanthology.org/nodalida-main-page/2020.acl-srw.10.pdf