mlsanityData to 7 Oct 2026

On Systematic Style Differences between Unsupervised and Supervised MT and an Application for High-Resource Machine Translation

arXiv 2106.15818 ↗First posted 30 Jun 2021cs.CLcited by 4

Not on the map. Why?

Abstract (arXiv)

Modern unsupervised machine translation (MT) systems reach reasonable translation quality under clean and controlled data conditions. As the performance gap between supervised and unsupervised MT narrows, it is interesting to ask whether the different training methods result in systematically different output beyond what is visible via quality metrics like adequacy or BLEU. We compare translations from supervised and unsupervised MT systems of similar quality, finding that unsupervised output is more fluent and more structurally different in comparison to human translation than is supervised MT. We then demonstrate a way to combine the benefits of both methods into a single system which results in improved adequacy and fluency as rated by human evaluators. Our results open the door to interesting discussions about how supervised and unsupervised MT might be different yet mutually-beneficial.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 27 of its 62 references that are in the map. First posted on arXiv 30 Jun 2021.

Builds on

27 of 62 references.

References · 27

  1. 2017Attention Is All You Needcited by 59,932
  2. 2018A Call for Clarity in Reporting BLEU Scorescited by 1,790
  3. 2019Cross-lingual Language Model Pretrainingcited by 1,088
  4. 2020Multilingual Denoising Pre-training for Neural Machine Translationcited by 1,028
  5. 2015Improving Neural Machine Translation Models with Monolingual Datacited by 784
  6. 2017Word Translation Without Parallel Datacited by 480
  7. 2019MASS: Masked Sequence to Sequence Pre-training for Language Generationcited by 345
  8. 2021Experts, Errors, and Context: A Large-Scale Study of Human Evaluation for Machine Translationcited by 333
  9. 2017Unsupervised Machine Translation Using Monolingual Corpora Onlycited by 272
  10. 2017Unsupervised Neural Machine Translationcited by 148
  11. 2018Phrase-Based & Neural Unsupervised Machine Translationcited by 138
  12. 2020BLEU might be Guilty but References are not Innocentcited by 109
15 more
  1. 2019Lingvo: a Modular and Scalable Framework for Sequence-to-Sequence Modelingcited by 86
  2. 2019Tagged Back-Translationcited by 76
  3. 2021Machine Translationese: Effects of Algorithmic Bias on Linguistic Complexity in Machine Translationcited by 70
  4. 2018Attaining the Unattainable? Reassessing Claims of Human Parity in Neural Machine Translationcited by 58
  5. 2020Leveraging Monolingual Data with Self-Supervision for Multilingual Neural Machine Translationcited by 51
  6. 2019The Effect of Translationese in Machine Translation Test Setscited by 49
  7. 2019Lost in Translation: Loss and Decay of Linguistic Richness in Machine Translationcited by 45
  8. 2019On The Evaluation of Machine Translation Systems Trained With Back-Translationcited by 40
  9. 2019Translationese in Machine Translation Evaluationcited by 38
  10. 2019An Effective Approach to Unsupervised Machine Translationcited by 37
  11. 2019Translationese as a Language in "Multilingual" NMTcited by 36
  12. 2019Post-editese: an Exacerbated Translationesecited by 29
  13. 2019Domain, Translationese and Noise in Synthetic Data for Neural Machine Translationcited by 26
  14. 2020Reassessing Claims of Human Parity and Super-Human Performance in Machine Translation at WMT 2019cited by 20
  15. 2015A Survey of Word Reordering in Statistical Machine Translation: Computational Models and Language Phenomenacited by 5

Built on by

Cited by year

Citing papers per year: 2022: 2; 2023: 1; 2024: 1.

4 papers.

  1. 2022Building Machine Translation Systems for the Next Thousand Languagescited by 69
  2. 2024To Diverge or Not to Diverge: A Morphosyntactic Perspective on Machine Translation vs Human Translationcited by 10
  3. 2022Improving Multilingual Neural Machine Translation System for Indic Languagescited by 6
  4. 2023An Empirical study of Unsupervised Neural Machine Translation: analyzing NMT output, model's behavior and sentences' contributioncited by 0