# Explaining and Generalizing Back-Translation through Wake-Sleep

arXiv: https://arxiv.org/abs/1806.04402 · HTML: https://mlsanity.com/paper/1806.04402/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Jun 2018.

## Built on by (most cited first, 7 of 7)

- [Neural Machine Translation for Low-Resource Languages: A Survey](https://mlsanity.com/paper/2106.15115/) (2021)
- [Unsupervised Neural Machine Translation with Generative Language Models Only](https://mlsanity.com/paper/2110.05448/) (2021)
- [HintedBT: Augmenting Back-Translation with Quality and Transliteration Hints](https://mlsanity.com/paper/2109.04443/) (2021)
- [The Impact of Syntactic and Semantic Proximity on Machine Translation with Back-Translation](https://mlsanity.com/paper/2403.18031/) (2024)
- [Learning from Multiple Sources for Data-to-Text and Text-to-Data](https://mlsanity.com/paper/2302.11269/) (2023)
- [Shared Latent Space by Both Languages in Non-Autoregressive Neural Machine Translation](https://mlsanity.com/paper/2305.03511/) (2023)
- [End-to-End Training for Back-Translation with Categorical Reparameterization Trick](https://mlsanity.com/paper/2202.08465/) (2022)

## Abstract (arXiv)

Back-translation has become a commonly employed heuristic for semi-supervised neural machine translation. The technique is both straightforward to apply and has led to state-of-the-art results. In this work, we offer a principled interpretation of back-translation as approximate inference in a generative model of bitext and show how the standard implementation of back-translation corresponds to a single iteration of the wake-sleep algorithm in our proposed model. Moreover, this interpretation suggests a natural iterative generalization, which we demonstrate leads to further improvement of up to 1.6 BLEU.