Copied Monolingual Data Improves Low-Resource Neural Machine Translation
Open accessThe DOI matches its Crossref record; 18 of the 18 references it cites were found in a registry or an index.
Abstract
We train a neural machine translation (NMT) system to both translate sourcelanguage text and copy target-language text, thereby exploiting monolingual corpora in the target language.Specifically, we create a bitext from the monolingual text in the target language so that each source sentence is identical to the target sentence.This copied data is then mixed with the parallel corpus and the NMT system is trained like normal, with no metadata to distinguish the two input languages.Our proposed method proves to be an effective way of incorporating monolingual data into low-resource NMT.On Turkish↔English and Romanian↔English translation tasks, we see gains of up to 1.2 BLEU over a strong baseline with back-translation.Further analysis shows that the linguistic phenomena behind these gains are different from and largely orthogonal to back-translation, with our copied corpus method improving accuracy on named entities and other words that should remain identical between the source and target languages.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946 · cited 2×
“The models are trained using Adam (Kingma and Ba, 2015) with a minibatch size of 80 and a maximum Language pair Parallel Monolingual EN↔TR 207 373 414 746 EN↔RO 608 320 608 320 EN↔DE 5 852 458 10 000 000”
MethodLearning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation
Cho, K., van Merrienboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., & Bengio, Y. (2014) · Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.3115/v1/d14-1179 · Cited by 25,545 · cited 2×
“We evaluate our approach using sequence-to-sequence neural machine translation (Cho et al., 2014; Kalchbrenner and Blunsom, 2013; Sutskever et al., 2014) augmented with attention (Bahdanau et al., 2015).”
MethodBLEU: a method for automatic evaluation of machine translation
Papineni, K., Roukos, S., Ward, T., & Zhu, W.-J. (2001) · Proceedings of the 40th Annual Meeting on Association for Computational Linguistics - ACL '02 · doi:10.3115/1073083.1073135 · Cited by 21,344 · cited 2×
“For each model, we report case-sensitive detokenized BLEU (Papineni et al., 2002) calculated using mteval-v13a.”
MethodSequence to Sequence Learning with Neural Networks
Sutskever, I., Vinyals, O., & Le, Q. V. (2014) · arXiv · doi:10.48550/arxiv.1409.3215 · Cited by 17,766 · cited 2×
“We evaluate our approach using sequence-to-sequence neural machine translation (Cho et al., 2014; Kalchbrenner and Blunsom, 2013; Sutskever et al., 2014) augmented with attention (Bahdanau et al., 2015).”
MethodNeural Machine Translation by Jointly Learning to Align and Translate
Bahdanau, D., Cho, K., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1409.0473 · Cited by 16,213 · cited 3×
“We evaluate our approach using sequence-to-sequence neural machine translation (Cho et al., 2014; Kalchbrenner and Blunsom, 2013; Sutskever et al., 2014) augmented with attention (Bahdanau et al., 2015).”
MethodGoogle’s Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation
Johnson, M., Schuster, M., Le, Q. V., Krikun, M., Wu, Y., Chen, Z., Thorat, N., Viégas, F., Wattenberg, M., Corrado, G., et al. (2017) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00065 · Cited by 1,747 · cited 1×
“Further work used the same encoder and decoder for multi-way translation (Johnson et al., 2016).”
MethodA theoretically grounded application of dropout in recurrent neural networks
Gal, Y., & Ghahramani, Z. (2016) · Cambridge University Engineering Department Publications Database · Cited by 1,600
Recurrent Continuous Translation Models
Kalchbrenner, N., & Blunsom, P. (2013) · Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d13-1176 · Cited by 1,232 · cited 1×
“We evaluate our approach using sequenceto-sequence neural machine translation (Cho et al., 2014; Kalchbrenner and Blunsom, 2013; Sutskever et al., 2014) augmented with attention (Bahdanau et al.”
MethodKenLM: Faster and Smaller Language Model Queries
Heafield, K. (2011) · Cited by 1,145
Self-citationA Theoretically Grounded Application of Dropout in Recurrent Neural Networks
Gal, Y., & Ghahramani, Z. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1512.05287 · Cited by 766 · cited 1×
“We apply dropout (Gal and Ghahramani, 2016) in all of our EN↔TR and EN↔RO systems with a probability of 0.”
MethodDual Learning for Machine Translation
Xia, Y., He, D., Qin, T., Wang, L., Yu, N., Liu, T., & Ma, W. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1611.00179 · Cited by 636
Findings of the 2016 Conference on Machine Translation
Bojar, O., Chatterjee, R., Federmann, C., Graham, Y., Haddow, B., Huck, M., Jimeno Yepes, A., Koehn, P., Logacheva, V., Monz, C., et al. (2016) · Proceedings of the First Conference on Machine Translation: Volume 2, Shared Task Papers · doi:10.18653/v1/w16-2301 · Cited by 539
On Using Monolingual Corpora in Neural Machine Translation
Gülçehre, Ç., Fırat, O., Xu, K., Cho, K., Barrault, L., Lin, H.-C., Bougares, F., Schwenk, H., & Bengio, Y. (2015) · HAL (Le Centre pour la Communication Scientifique Directe) · doi:10.48550/arxiv.1503.03535 · Cited by 466 · cited 4×
“Since this back-translation method outperforms previous methods that only train the decoder (Gulcehre et al., 2015; Sennrich et al., 2016b), we use”
BackgroundMethodInfluentialNematus: a Toolkit for Neural Machine Translation
Sennrich, R., Firat, O., Cho, K., Birch, A., Haddow, B., Hitschler, J., Junczys-Dowmunt, M., Läubli, S., Miceli Barone, A. V., Mokry, J., et al. (2017) · Proceedings of the Software Demonstrations of the 15th Conference of the European Chapter of the Association for Computational Linguistics · doi:10.18653/v1/e17-3017 · Cited by 291
Self-citationExploiting Source-side Monolingual Data in Neural Machine Translation
Zhang, J., & Zong, C. (2016) · Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d16-1160 · Cited by 277
Montreal Neural Machine Translation Systems for WMT’15
Jean, S., Firat, O., Cho, K., Memisevic, R., & Bengio, Y. (2015) · Proceedings of the Tenth Workshop on Statistical Machine Translation · doi:10.18653/v1/w15-3014 · Cited by 134 · cited 1×
“Jean et al. (2015) and Gulcehre et al. (2015) used a 5 -gram language model and a recurrent neural network language model (RNNLM), respectively, to re-rank NMT outputs.”
MethodEdinburgh Neural Machine Translation Systems for WMT 16
Sennrich, R., Haddow, B., & Birch, A. (2016) · Proceedings of the First Conference on Machine Translation: Volume 2, Shared Task Papers · doi:10.18653/v1/w16-2323 · Cited by 112 · cited 7×
“Recent work (Gul-cehre et al., 2015; Sennrich et al., 2016b) has investigated incorporating monolingual training data (particularly on the target side) into NMT.”
BackgroundMethodInfluentialSemi-Supervised Learning for Neural Machine TranslationPreprint
Cheng, Y., Xu, W. H., He, Z., He, W. P., Wu, H., Sun, M., & Liu, Y. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1606.04596 · Cited by 92 · cited 2×
“Cheng et al. (2016) accomplished this by concatenating source → target and target → source NMT systems to create an autoen-coder.”
Method
Cited by
Understanding Back-Translation at Scale
Edunov, S., Ott, M., Auli, M., & Grangier, D. (2018) · Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d18-1045 · Cited by 993
Unsupervised Neural Machine Translation
Artetxe, M., Labaka, G., Agirre, E., & Cho, K. (2017) · arXiv · doi:10.48550/arxiv.1710.11041 · Cited by 632
Neural Machine Translation: A Review
Stahlberg, F. (2020) · Journal of Artificial Intelligence Research · doi:10.1613/jair.1.12007 · Cited by 397
Neural Machine Translation for Low-resource Languages: A Survey
Ranathunga, S., Lee, E.-S. A., Prifti Skenduli, M., Shekhar, R., Alam, M., & Kaur, R. (2023) · ACM Computing Surveys · doi:10.1145/3567592 · Cited by 313
Iterative Back-Translation for Neural Machine Translation
Hoang, V. C. D., Koehn, P., Haffari, G., & Cohn, T. (2018) · Proceedings of the 2nd Workshop on Neural Machine Translation and Generation · doi:10.18653/v1/w18-2703 · Cited by 235
“Currey et al. (2017) reported similar findings for low resource conditions, showing that even poor translations can be beneficial.”
Background
Metadata
- Authors
- Currey, A., Miceli Barone, A. V., & Heafield, K.
- Venue
- Proceedings of the Second Conference on Machine Translation
- Published
- 2017
- Pages
- 148-156
- OpenAlex
- W2756566411
- Cited by
- 208
- References
- 18
- FWCI
- 15.88
- Type
- conference paper
- Language
- en
- Open access
- hybrid
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.