Search Engine Guided Non-Parametric Neural Machine Translation
This work addresses translation quality in NMT by leveraging external data retrieval, offering a domain-specific improvement for machine translation systems.
The paper tackled the problem of neural machine translation by enabling the model to access the entire training set during inference, using a search engine to retrieve relevant sentence pairs and a novel translation memory enhanced NMT model. The result was significant performance improvements over baselines on three language pairs, with gains increasing when more relevant pairs were retrieved.
In this paper, we extend an attention-based neural machine translation (NMT) model by allowing it to access an entire training set of parallel sentence pairs even after training. The proposed approach consists of two stages. In the first stage--retrieval stage--, an off-the-shelf, black-box search engine is used to retrieve a small subset of sentence pairs from a training set given a source sentence. These pairs are further filtered based on a fuzzy matching score based on edit distance. In the second stage--translation stage--, a novel translation model, called translation memory enhanced NMT (TM-NMT), seamlessly uses both the source sentence and a set of retrieved sentence pairs to perform the translation. Empirical evaluation on three language pairs (En-Fr, En-De, and En-Es) shows that the proposed approach significantly outperforms the baseline approach and the improvement is more significant when more relevant sentence pairs were retrieved.