# Bi-Directional Differentiable Input Reconstruction for Low-Resource Neural Machine Translation

arXiv: https://arxiv.org/abs/1811.01116 · HTML: https://mlsanity.com/paper/1811.01116/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Nov 2018.

## Built on by (most cited first, 4 of 4)

- [Survey of Low-Resource Machine Translation](https://mlsanity.com/paper/2109.00486/) (2021)
- [How Large Language Models are Transforming Machine-Paraphrased Plagiarism](https://mlsanity.com/paper/2210.03568/) (2022)
- [Learning to translate by learning to communicate](https://mlsanity.com/paper/2207.07025/) (2022)
- [Toward Improving Coherence and Diversity of Slogan Generation](https://mlsanity.com/paper/2102.05924/) (2021)

## Abstract (arXiv)

We aim to better exploit the limited amounts of parallel text available in low-resource settings by introducing a differentiable reconstruction loss for neural machine translation (NMT). This loss compares original inputs to reconstructed inputs, obtained by back-translating translation hypotheses into the input language. We leverage differentiable sampling and bi-directional NMT to train models end-to-end, without introducing additional parameters. This approach achieves small but consistent BLEU improvements on four language pairs in both translation directions, and outperforms an alternative differentiable reconstruction strategy based on hidden states.