# Learning to Reuse Translations: Guiding Neural Machine Translation with Examples

arXiv: https://arxiv.org/abs/1911.10732 · HTML: https://mlsanity.com/paper/1911.10732/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Nov 2019.

## Built on by (most cited first, 5 of 5)

- [A Survey on Retrieval-Augmented Text Generation](https://mlsanity.com/paper/2202.01110/) (2022)
- [Neural Machine Translation with Monolingual Translation Memory](https://mlsanity.com/paper/2105.11269/) (2021)
- [Non-Parametric Online Learning from Human Feedback for Neural Machine Translation](https://mlsanity.com/paper/2109.11136/) (2021)
- [Prompting Neural Machine Translation with Translation Memories](https://mlsanity.com/paper/2301.05380/) (2023)
- [Rethinking Translation Memory Augmented Neural Machine Translation](https://mlsanity.com/paper/2306.06948/) (2023)

## Abstract (arXiv)

In this paper, we study the problem of enabling neural machine translation (NMT) to reuse previous translations from similar examples in target prediction. Distinguishing reusable translations from noisy segments and learning to reuse them in NMT are non-trivial. To solve these challenges, we propose an Example-Guided NMT (EGNMT) framework with two models: (1) a noise-masked encoder model that masks out noisy words according to word alignments and encodes the noise-masked sentences with an additional example encoder and (2) an auxiliary decoder model that predicts reusable words via an auxiliary decoder sharing parameters with the primary decoder. We define and implement the two models with the state-of-the-art Transformer. Experiments show that the noise-masked encoder model allows NMT to learn useful information from examples with low fuzzy match scores (FMS) while the auxiliary decoder model is good for high-FMS examples. More experiments on Chinese-English, English-German and English-Spanish translation demonstrate that the combination of the two EGNMT models can achieve improvements of up to +9 BLEU points over the baseline system and +7 BLEU points over a two-encoder Transformer.