mlsanityData to 7 Oct 2026

When and Why is Unsupervised Neural Machine Translation Useless?

arXiv 2004.10581 ↗First posted 22 Apr 2020cs.CLcited by 34

Not on the map. Why?

Abstract (arXiv)

This paper studies the practicality of the current state-of-the-art unsupervised methods in neural machine translation (NMT). In ten translation tasks with various data settings, we analyze the conditions under which the unsupervised methods fail to produce reasonable translations. We show that their performance is severely affected by linguistic dissimilarity and domain mismatch between source and target monolingual data. Such conditions are common for low-resource language pairs, where unsupervised learning works poorly. In all of our experiments, supervised and semi-supervised baselines with 50k-sentence bilingual data outperform the best unsupervised results. Our analyses pinpoint the limits of the current unsupervised NMT and also suggest immediate research directions.

Key facts. Cited by 34 papers in the map (data through 7 Oct 2026). First posted on arXiv 22 Apr 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 14; 2022: 7; 2023: 6; 2024: 3; 2025: 2; 2026: 2.

34 papers.

Sort:

By area:

  1. 2021Neural Machine Translation for Low-Resource Languages: A Surveycited by 108
  2. 2021Survey of Low-Resource Machine Translationcited by 92
  3. 2023The unreasonable effectiveness of few-shot learning for machine translationcited by 91
  4. 2022Towards the Next 1000 Languages in Multilingual Machine Translation: Exploring the Synergy Between Supervised and Self-Supervised Learningcited by 19
  5. 2021Crossing the Conversational Chasm: A Primer on Natural Language Processing for Multilingual Task-Oriented Dialogue Systemscited by 16
  6. 2021Towards Making the Most of Multilingual Pretraining for Zero-Shot Neural Machine Translationcited by 8
  7. 2021Towards Universality in Multilingual Text Rewritingcited by 7
  8. 2022IsoVec: Controlling the Relative Isomorphism of Word Embedding Spacescited by 7
  9. 2021Summarising Historical Text in Modern Languagescited by 6
  10. 2022A Theory of Unsupervised Translation Motivated by Understanding Animal Communicationcited by 6
  11. 2021Majority Voting with Bidirectional Pre-translation For Bitext Retrievalcited by 5
  12. 2021"Wikily" Supervised Neural Translation Tailored to Cross-Lingual Taskscited by 5
22 more
  1. 2023Translating away Translationese without Parallel Datacited by 5
  2. 2021Unsupervised Machine Translation On Dravidian Languagescited by 4
  3. 2021Integrating Unsupervised Data Generation into Self-Supervised Neural Machine Translation for Low-Resource Languagescited by 4
  4. 2021Low-Resource Machine Translation Training Curriculum Fit for Low-Resource Languagescited by 3
  5. 2021Crosslingual Embeddings are Essential in UNMT for Distant Languages: An English to IndoAryan Case Studycited by 3
  6. 2022Multilingual Bidirectional Unsupervised Translation Through Multilingual Finetuning and Back-Translationcited by 3
  7. 2023GRI: Graph-based Relative Isomorphism of Word Embedding Spacescited by 3
  8. 2022Cost-Effective Training in Low-Resource Neural Machine Translationcited by 2
  9. 2022Exploiting Social Media Content for Self-Supervised Style Transfercited by 2
  10. 2024The Impact of Syntactic and Semantic Proximity on Machine Translation with Back-Translationcited by 2
  11. 2021Self-Learning for Zero Shot Neural Machine Translationcited by 1
  12. 2021Using Document Similarity Methods to create Parallel Datasets for Code Translationcited by 1
  13. 2022Domain Mismatch Doesn't Always Prevent Cross-Lingual Transfer Learningcited by 1
  14. 2023On the Copying Problem of Unsupervised NMT: A Training Schedule with a Language Discriminator Losscited by 1
  15. 2024Pointer-Generator Networks for Low-Resource Machine Translation: Don't Copy That!cited by 1
  16. 2023Relevance-guided Neural Machine Translationcited by 0
  17. 2023An Empirical study of Unsupervised Neural Machine Translation: analyzing NMT output, model's behavior and sentences' contributioncited by 0
  18. 2024Decoupled Vocabulary Learning Enables Zero-Shot Translation from Unseen Languagescited by 0
  19. 2025Mind the Language Gap in Digital Humanities: LLM-Aided Translation of SKOS Thesauricited by 0
  20. 2025On Non-interactive Evaluation of Animal Communication Translatorscited by 0
  21. 2026Ensemble Self-Training for Unsupervised Machine Translationcited by 0
  22. 2026Evaluating In-Context Translation with Synchronous Context-Free Grammar Transductioncited by 0