Lost in Machine Translation: A Method to Reduce Meaning Loss

Reuben Cohn-Gordon, Noah Goodman


Abstract
A desideratum of high-quality translation systems is that they preserve meaning, in the sense that two sentences with different meanings should not translate to one and the same sentence in another language. However, state-of-the-art systems often fail in this regard, particularly in cases where the source and target languages partition the “meaning space” in different ways. For instance, “I cut my finger.” and “I cut my finger off.” describe different states of the world but are translated to French (by both Fairseq and Google Translate) as “Je me suis coupé le doigt.”, which is ambiguous as to whether the finger is detached. More generally, translation systems are typically many-to-one (non-injective) functions from source to target language, which in many cases results in important distinctions in meaning being lost in translation. Building on Bayesian models of informative utterance production, we present a method to define a less ambiguous translation system in terms of an underlying pre-trained neural sequence-to-sequence model. This method increases injectivity, resulting in greater preservation of meaning as measured by improvement in cycle-consistency, without impeding translation quality (measured by BLEU score).
Anthology ID:
N19-1042
Volume:
Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)
Month:
June
Year:
2019
Address:
Minneapolis, Minnesota
Editors:
Jill Burstein, Christy Doran, Thamar Solorio
Venue:
NAACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
437–441
Language:
URL:
https://aclanthology.org/N19-1042
DOI:
10.18653/v1/N19-1042
Bibkey:
Cite (ACL):
Reuben Cohn-Gordon and Noah Goodman. 2019. Lost in Machine Translation: A Method to Reduce Meaning Loss. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pages 437–441, Minneapolis, Minnesota. Association for Computational Linguistics.
Cite (Informal):
Lost in Machine Translation: A Method to Reduce Meaning Loss (Cohn-Gordon & Goodman, NAACL 2019)
Copy Citation:
PDF:
https://aclanthology.org/N19-1042.pdf
Code
 reubenharry/pragmatic-translation