mlsanityData to 14 Oct 2026

Joint Dropout: Improving Generalizability in Low-Resource Neural Machine Translation through Phrase Pair Variables

arXiv 2307.12835 ↗First posted 24 Jul 2023cs.CLcited by 1

Not on the map. Why?

Abstract (arXiv)

Despite the tremendous success of Neural Machine Translation (NMT), its performance on low-resource language pairs still remains subpar, partly due to the limited ability to handle previously unseen inputs, i.e., generalization. In this paper, we propose a method called Joint Dropout, that addresses the challenge of low-resource neural machine translation by substituting phrases with variables, resulting in significant enhancement of compositionality, which is a key aspect of generalization. We observe a substantial improvement in translation quality for language pairs with minimal resources, as seen in BLEU and Direct Assessment scores. Furthermore, we conduct an error analysis, and find Joint Dropout to also enhance generalizability of low-resource NMT in terms of robustness and adaptability across different domains

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 27 of its 42 references that are in the map. First posted on arXiv 24 Jul 2023.

Builds on

27 of 42 references.

References · 27

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Neural Machine Translation of Rare Words with Subword Unitscited by 2,918
  3. 2018A Call for Clarity in Reporting BLEU Scorescited by 1,799
  4. 2019fairseq: A Fast, Extensible Toolkit for Sequence Modelingcited by 1,488
  5. 2020Multilingual Denoising Pre-training for Neural Machine Translationcited by 1,035
  6. 2021The FLORES-101 Evaluation Benchmark for Low-Resource and Multilingual Machine Translationcited by 560
  7. 2017Six Challenges for Neural Machine Translationcited by 335
  8. 2017Synthetic and Natural Noise Both Break Neural Machine Translationcited by 306
  9. 2019Measuring Compositional Generalization: A Comprehensive Method on Realistic Datacited by 284
  10. 2019Compositionality decomposed: how do neural networks generalise?cited by 265
  11. 2020COGS: A Compositional Generalization Challenge Based on Semantic Interpretationcited by 263
  12. 2018When and Why are Pre-trained Word Embeddings Useful for Neural Machine Translation?cited by 126
15 more
  1. 2019Robust Neural Machine Translation with Doubly Adversarial Inputscited by 106
  2. 2022State-of-the-art generalisation research in NLP: A taxonomy and reviewcited by 98
  3. 2016Edinburgh Neural Machine Translation Systems for WMT 16cited by 76
  4. 2018SwitchOut: an Efficient Data Augmentation Algorithm for Neural Machine Translationcited by 72
  5. 2021The paradox of the compositionality of natural language: a neural machine translation case studycited by 59
  6. 2018MTNT: A Testbed for Machine Translation of Noisy Textcited by 54
  7. 2021Improving Compositional Generalization with Latent Structure and Data Augmentationcited by 46
  8. 2021On Compositional Generalization of Neural Machine Translationcited by 34
  9. 2020Optimizing Transformer for Low-Resource Neural Machine Translationcited by 24
  10. 2021Improving Compositional Generalization with Self-Training for Data-to-Text Generationcited by 19
  11. 2021Improving Compositional Generalization in Classification Tasks via Structure Annotationscited by 12
  12. 2017A Multifaceted Evaluation of Neural versus Phrase-Based Machine Translation for 9 Language Directionscited by 10
  13. 2020The Unreasonable Volatility of Neural Machine Translation Modelscited by 9
  14. 2020Token Drop mechanism for Neural Machine Translationcited by 8
  15. 2022Improving Compositional Generalization in Math Word Problem Solvingcited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Semantically Orthogonal Framework for Citation Classification: Disentangling Intent and Contentcited by 0