mlsanityData to 7 Oct 2026

Meta-Learning for Low-Resource Neural Machine Translation

arXiv 1808.08437 ↗First posted 25 Aug 2018cs.CL cs.LGcited by 105

Not on the map. Why?

Abstract (arXiv)

In this paper, we propose to extend the recently introduced model-agnostic meta-learning algorithm (MAML) for low-resource neural machine translation (NMT). We frame low-resource translation as a meta-learning problem, and we learn to adapt to low-resource languages based on multilingual high-resource language tasks. We use the universal lexical representation~citepgu2018universal to overcome the input-output mismatch across different languages. We evaluate the proposed meta-learning strategy using eighteen European languages (Bg, Cs, Da, De, El, Es, Et, Fr, Hu, It, Lt, Nl, Pl, Pt, Sk, Sl, Sv and Ru) as source tasks and five diverse languages (Ro, Lv, Fi, Tr and Ko) as target tasks. We show that the proposed approach significantly outperforms the multilingual, transfer learning based approach~citepzoph2016transfer and enables us to train a competitive NMT system with only a fraction of training examples. For instance, the proposed approach can achieve as high as 22.04 BLEU on Romanian-English WMT'16 by seeing only 16,000 translated words (~600 parallel sentences).

Key facts. Cited by 105 papers in the map (data through 7 Oct 2026). First posted on arXiv 25 Aug 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 45; 2022: 24; 2023: 20; 2024: 9; 2025: 4; 2026: 3.

105 papers.

Sort:

By area:

  1. 2023Knowledge Editing for Large Language Models: A Surveycited by 189
  2. 2021Meta-learning via Language Model In-context Tuningcited by 163
  3. 2021CrossFit: A Few-shot Learning Challenge for Cross-task Generalization in NLPcited by 142
  4. 2021Neural Machine Translation for Low-Resource Languages: A Surveycited by 108
  5. 2021Survey of Low-Resource Machine Translationcited by 92
  6. 2022Few-Shot Class-Incremental Learning by Sampling Multi-Phase Taskscited by 87
  7. 2021FLEX: Unifying Evaluation for Few-Shot NLPcited by 67
  8. 2023Advances and Challenges in Meta-Learning: A Technical Reviewcited by 65
  9. 2021Meta-Learning to Compositionally Generalizecited by 62
  10. 2024(Perhaps) Beyond Human Translation: Harnessing Multi-Agent Collaboration for Translating Ultra-Long Literary Textscited by 49
  11. 2023STUNT: Few-shot Tabular Learning with Self-generated Tasks from Unlabeled Tablescited by 48
  12. 2021Controllable Semantic Parsing via Retrieval Augmentationcited by 35
38 more
  1. 2023A Survey on Knowledge Editing of Neural Networkscited by 34
  2. 2021Continual Mixed-Language Pre-Training for Extremely Low-Resource Neural Machine Translationcited by 29
  3. 2021Meta-Curriculum Learning for Domain Adaptation in Neural Machine Translationcited by 28
  4. 2021Cross-Lingual Abstractive Summarization with Limited Parallel Resourcescited by 27
  5. 2022Match the Script, Adapt if Multilingual: Analyzing the Effect of Multilingual Pretraining on Cross-lingual Transferabilitycited by 26
  6. 2022Meta Learning for Natural Language Processing: A Surveycited by 26
  7. 2023Meta-learning approaches for few-shot learning: A survey of recent advancescited by 25
  8. 2021Zero-shot Cross-lingual Transfer of Neural Machine Translation with Multilingual Pretrained Encoderscited by 24
  9. 2021The Case for Translation-Invariant Self-Attention in Transformer-Based Language Modelscited by 23
  10. 2021How to Train Your MAML to Excel in Few-Shot Classificationcited by 22
  11. 2021MetaXL: Meta Representation Transformation for Low-resource Cross-lingual Learningcited by 21
  12. 2021ProtAugment: Unsupervised diverse short-texts paraphrasing for intent detection meta-learningcited by 18
  13. 2021Question Answering Survey: Directions, Challenges, Datasets, Evaluation Matricescited by 17
  14. 2021Few Shot Dialogue State Tracking using Meta-learningcited by 16
  15. 2021Meta Back-translationcited by 16
  16. 2021Few-Shot Learning with a Strong Teachercited by 16
  17. 2021A Survey on Low-Resource Neural Machine Translationcited by 16
  18. 2023CrossCodeBench: Benchmarking Cross-Task Generalization of Source Code Modelscited by 16
  19. 2022Boosting Natural Language Generation from Instructions with Meta-Learningcited by 15
  20. 2021Multilingual and cross-lingual document classification: A meta-learning approachcited by 14
  21. 2021Meta-Transfer Learning for Low-Resource Abstractive Summarizationcited by 14
  22. 2021Meta-Learning for Effective Multi-task and Multilingual Modellingcited by 13
  23. 2021A Representation Learning Perspective on the Importance of Train-Validation Splitting in Meta-Learningcited by 12
  24. 2022Prompt-Based Metric Learning for Few-Shot NERcited by 12
  25. 2021Improving Both Domain Robustness and Domain Adaptability in Machine Translationcited by 11
  26. 2022Meta-Learning a Cross-lingual Manifold for Semantic Parsingcited by 11
  27. 2023Learning to Initialize: Can Meta Learning Improve Cross-task Generalization in Prompt Tuning?cited by 11
  28. 2023Model-Agnostic Meta-Learning for Multilingual Hate Speech Detectioncited by 11
  29. 2024GPTs Are Multilingual Annotators for Sequence Generation Taskscited by 11
  30. 2021X-METRA-ADA: Cross-lingual Meta-Transfer Learning Adaptation to Natural Language Understanding and Question Answeringcited by 10
  31. 2021kFolden: k-Fold Ensemble for Out-Of-Distribution Detectioncited by 10
  32. 2021Knowledge-Aware Meta-learning for Low-Resource Text Classificationcited by 10
  33. 2021Diverse Distributions of Self-Supervised Tasks for Meta-Learning in NLPcited by 10
  34. 2022Continued Pretraining for Better Zero- and Few-Shot Promptabilitycited by 10
  35. 2022Learning a Better Initialization for Soft Prompts via Meta-Learningcited by 9
  36. 2022Provable Generalization of Overparameterized Meta-learning Trained with SGDcited by 9
  37. 2021Few-shot learning through contextual data augmentationcited by 8
  38. 2022Improving Meta-learning for Low-resource Text Classification and Generation via Memory Imitationcited by 8

50 most cited of 105 All 105 (JSON)