Abstract
We present a simple method to improve neural translation of a low-resource language pair using parallel data from a related, also low-resource, language pair. The method is based on the transfer method of Zoph et al., but whereas their method ignores any source vocabulary overlap, ours exploits it. First, we split words using Byte Pair Encoding (BPE) to increase vocabulary overlap. Then, we train a model on the first language pair and transfer its parameters, including its source word embeddings, to another model and continue training on the second language pair. Our experiments show that transfer learning helps word-based translation only slightly, but when used on top of a much stronger BPE baseline, it yields larger improvements of up to 4.3 BLEU.- Anthology ID:
- I17-2050
- Volume:
- Proceedings of the Eighth International Joint Conference on Natural Language Processing (Volume 2: Short Papers)
- Month:
- November
- Year:
- 2017
- Address:
- Taipei, Taiwan
- Editors:
- Greg Kondrak, Taro Watanabe
- Venue:
- IJCNLP
- SIG:
- Publisher:
- Asian Federation of Natural Language Processing
- Note:
- Pages:
- 296–301
- Language:
- URL:
- https://aclanthology.org/I17-2050
- DOI:
- Cite (ACL):
- Toan Q. Nguyen and David Chiang. 2017. Transfer Learning across Low-Resource, Related Languages for Neural Machine Translation. In Proceedings of the Eighth International Joint Conference on Natural Language Processing (Volume 2: Short Papers), pages 296–301, Taipei, Taiwan. Asian Federation of Natural Language Processing.
- Cite (Informal):
- Transfer Learning across Low-Resource, Related Languages for Neural Machine Translation (Nguyen & Chiang, IJCNLP 2017)
- PDF:
- https://preview.aclanthology.org/naacl24-info/I17-2050.pdf