---
title: "Google's Multilingual Neural Machine Translation System:"
description: "The authors from Google address the task of translating between many languages with a single neural model (NMT). Cited by 883 papers here."
authors: ["Melvin Johnson", "Mike Schuster", "Quoc V. Le", "Maxim Krikun", "Yonghui Wu", "Zhifeng Chen", "Nikhil Thorat", "Fernanda Viégas", "Martin Wattenberg", "Greg Corrado", "Macduff Hughes", "Jeffrey Dean"]
year: 2017
venue: "Transactions of the Association for Computational Linguistics"
source: "https://aclanthology.org/Q17-1024/"
doi: "10.1162/tacl_a_00065"
arxiv: "1611.04558"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/johnson-2017-google-s-multilingual-neural-machine/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/johnson-2017-google-s-multilingual-neural-machine/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2017](https://md.research.elamur.ai/free/en/papers/2017/) / Google's Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation

# Google's Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation

**Authors:** [Melvin Johnson](https://md.research.elamur.ai/free/en/authors/melvin-johnson/), [Mike Schuster](https://md.research.elamur.ai/free/en/authors/mike-schuster/), [Quoc V. Le](https://md.research.elamur.ai/free/en/authors/quoc-viet-le/), [Maxim Krikun](https://md.research.elamur.ai/free/en/authors/maxim-krikun/), [Yonghui Wu](https://md.research.elamur.ai/free/en/authors/yonghui-wu/), [Zhifeng Chen](https://md.research.elamur.ai/free/en/authors/zhifeng-chen/), Nikhil Thorat, [Fernanda Viégas](https://md.research.elamur.ai/free/en/authors/fernanda-viegas/), [Martin Wattenberg](https://md.research.elamur.ai/free/en/authors/martin-wattenberg/), [Greg Corrado](https://md.research.elamur.ai/free/en/authors/greg-s-corrado/), [Macduff Hughes](https://md.research.elamur.ai/free/en/authors/macduff-hughes/) and [Jeffrey Dean](https://md.research.elamur.ai/free/en/authors/jeff-dean/) · **Year:** 2017 · **Venue:** [Transactions of the Association for Computational Linguistics](https://md.research.elamur.ai/free/en/venues/tacl/) ([Transactions of the Association for Computational Linguistics (TACL) 2017](https://md.research.elamur.ai/free/en/venues/tacl/2017/)) · **Publisher:** [The MIT Press](https://md.research.elamur.ai/free/en/publishers/the-mit-press/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/Q17-1024/) · [PDF](https://aclanthology.org/Q17-1024.pdf) · [DOI 10.1162/tacl\_a\_00065](https://doi.org/10.1162/tacl_a_00065) · [arXiv 1611.04558](https://arxiv.org/abs/1611.04558) · [OpenAlex](https://openalex.org/W2550821151)

**Citations:** 1,747 citations (OpenAlex, 2026-10-14) · [883](https://md.research.elamur.ai/free/en/papers/johnson-2017-google-s-multilingual-neural-machine/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
The authors from Google address the task of translating between many languages with a single neural model (NMT). Previously, a separate system was usually trained for each language pair. They left the GNMT architecture unchanged and modified only the input data: an artificial token indicating the target language, for example `<2es>`, is added to the beginning of the sentence. The model identifies the source language on its own. The encoder, decoder, attention mechanism, and wordpiece vocabulary are shared across all languages. The model was trained on mixed multilingual data from the WMT benchmarks and on Google's production corpora. A single such model performs on par with or better than the best systems of its time on WMT'14/WMT'15, and low-resource languages benefit from joint training with high-resource languages. Most importantly, the model can translate between language pairs it did not see during training (zero-shot translation). This approach simplifies the deployment of hundreds of language directions. It also shows that cross-lingual knowledge transfer is possible in neural translation and that a shared cross-lingual representation (interlingua) is likely formed.

## Key points
- Method: the only change is a target-language token at the input. The model uses a shared wordpiece vocabulary, typically of 32,000 units (similar to BPE). Implemented in TensorFlow on top of GNMT.
- Results on WMT'14: for English→French the quality is comparable to the best systems, and for English→German it is better. For French→English (WMT'14) and German→English (WMT'15), the multilingual model also surpasses the then state of the art.
- On Google's production corpora, the models cover up to 12 language pairs in a single network. Many individual pairs are translated better as a result.
- Zero-shot: a model trained on Portuguese→English and English→Spanish produces acceptable Portuguese→Spanish translation. A small amount of data for this pair noticeably improves quality.
- t-SNE analysis showed that sentences with the same meaning in different languages group into shared clusters, which supports the interlingua hypothesis. The farther a zero-shot translation is from the "learned" representation, the lower the BLEU (Pearson correlation −0.42). The model also handles language switching within an input sentence (Japanese + Korean). When target-language tokens are mixed, the output usually switches from one language to the other at a weight of about 0.5, rather than producing a "hybrid" language.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [Google](https://md.research.elamur.ai/free/en/companies/google/) — Company, Mountain View, United States · 430 works · 424,363 citations

📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works

🧭 Topics (2): [🔁 Machine translation](https://md.research.elamur.ai/free/en/topics/machine-translation/) · 1,087 works · [🌍 Multilingual & cross-lingual NLP](https://md.research.elamur.ai/free/en/topics/multilingual-nlp/) · 828 works

💡 Concepts (1): [💡 Few-shot learning](https://md.research.elamur.ai/free/en/concepts/few-shot-learning/) · 45 works

## Cite

Cite as: Melvin Johnson, Mike Schuster, Quoc V. Le, Maxim Krikun, Yonghui Wu, Zhifeng Chen, Nikhil Thorat, Fernanda Viégas, Martin Wattenberg, Greg Corrado, Macduff Hughes and Jeffrey Dean. 2017. Google's Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation. Transactions of the Association for Computational Linguistics.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@article{johnson2017google,
  title = {{Google's Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation}},
  author = {Melvin Johnson, Mike Schuster, Quoc V. Le, Maxim Krikun, Yonghui Wu, Zhifeng Chen, Nikhil Thorat, Fernanda Viégas, Martin Wattenberg, Greg Corrado, Macduff Hughes and Jeffrey Dean},
  journal = {Transactions of the Association for Computational Linguistics},
  year = {2017},
  doi = {10.1162/tacl\_a\_00065},
  eprint = {1611.04558},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.1162/tacl_a_00065}
}
```

- [← previous What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?](https://md.research.elamur.ai/free/en/papers/kendall-2017-uncertainties-need-bayesian-deep-learning/)
- [next → An Overview of Multi-Task Learning in Deep Neural Networks](https://md.research.elamur.ai/free/en/papers/ruder-2017-overview-multi-task-learning-deep/)
- [↑ back to papers of 2017](https://md.research.elamur.ai/free/en/papers/2017/)

*Updated 29 October 2026*