Iterative Batch Back-Translation for Neural Machine Translation: A Conceptual Model

TL;DR

This paper introduces an iterative batch back-translation method for neural machine translation that improves data utilization and translation quality by systematically enhancing training data through multiple iterations.

Contribution

It proposes a novel conceptual model for iterative batch back-translation, enabling more efficient use of monolingual data in NMT training.

Findings

Outperforms standard back-translation in some language pairs

Enhances the quality of back-translations over iterations

Facilitates better utilization of monolingual data

Abstract

An effective method to generate a large number of parallel sentences for training improved neural machine translation (NMT) systems is the use of back-translations of the target-side monolingual data. Recently, iterative back-translation has been shown to outperform standard back-translation albeit on some language pairs. This work proposes the iterative batch back-translation that is aimed at enhancing the standard iterative back-translation and enabling the efficient utilization of more monolingual data. After each iteration, improved back-translations of new sentences are added to the parallel data that will be used to train the final forward model. The work presents a conceptual model of the proposed approach.