Abstract
Advanced neural machine translation (NMT) models generally implement encoder and decoder as multiple layers, which allows systems to model complex functions and capture complicated linguistic structures. However, only the top layers of encoder and decoder are leveraged in the subsequent process, which misses the opportunity to exploit the useful information embedded in other layers. In this work, we propose to simultaneously expose all of these signals with layer aggregation and multi-layer attention mechanisms. In addition, we introduce an auxiliary regularization term to encourage different layers to capture diverse information. Experimental results on widely-used WMT14 English-German and WMT17 Chinese-English translation data demonstrate the effectiveness and universality of the proposed approach.- Anthology ID:
- D18-1457
- Volume:
- Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing
- Month:
- October-November
- Year:
- 2018
- Address:
- Brussels, Belgium
- Venue:
- EMNLP
- SIG:
- SIGDAT
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 4253–4262
- Language:
- URL:
- https://aclanthology.org/D18-1457
- DOI:
- 10.18653/v1/D18-1457
- Cite (ACL):
- Zi-Yi Dou, Zhaopeng Tu, Xing Wang, Shuming Shi, and Tong Zhang. 2018. Exploiting Deep Representations for Neural Machine Translation. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 4253–4262, Brussels, Belgium. Association for Computational Linguistics.
- Cite (Informal):
- Exploiting Deep Representations for Neural Machine Translation (Dou et al., EMNLP 2018)
- PDF:
- https://preview.aclanthology.org/starsem-semeval-split/D18-1457.pdf