It’s Easier to Translate out of English than into it: Measuring Neural Translation Difficulty by Cross-Mutual Information
Emanuele Bugliarello, Sabrina J. Mielke, Antonios Anastasopoulos, Ryan Cotterell, Naoaki Okazaki
Abstract
The performance of neural machine translation systems is commonly evaluated in terms of BLEU. However, due to its reliance on target language properties and generation, the BLEU metric does not allow an assessment of which translation directions are more difficult to model. In this paper, we propose cross-mutual information (XMI): an asymmetric information-theoretic metric of machine translation difficulty that exploits the probabilistic nature of most neural machine translation models. XMI allows us to better evaluate the difficulty of translating text into the target language while controlling for the difficulty of the target-side generation component independent of the translation task. We then present the first systematic and controlled study of cross-lingual translation difficulties using modern neural translation systems. Code for replicating our experiments is available online at https://github.com/e-bug/nmt-difficulty.- Anthology ID:
- 2020.acl-main.149
- Volume:
- Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics
- Month:
- July
- Year:
- 2020
- Address:
- Online
- Editors:
- Dan Jurafsky, Joyce Chai, Natalie Schluter, Joel Tetreault
- Venue:
- ACL
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 1640–1649
- Language:
- URL:
- https://preview.aclanthology.org/build-pipeline-with-new-library/2020.acl-main.149/
- DOI:
- 10.18653/v1/2020.acl-main.149
- Cite (ACL):
- Emanuele Bugliarello, Sabrina J. Mielke, Antonios Anastasopoulos, Ryan Cotterell, and Naoaki Okazaki. 2020. It’s Easier to Translate out of English than into it: Measuring Neural Translation Difficulty by Cross-Mutual Information. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 1640–1649, Online. Association for Computational Linguistics.
- Cite (Informal):
- It’s Easier to Translate out of English than into it: Measuring Neural Translation Difficulty by Cross-Mutual Information (Bugliarello et al., ACL 2020)
- PDF:
- https://preview.aclanthology.org/build-pipeline-with-new-library/2020.acl-main.149.pdf
- Code
- e-bug/nmt-difficulty