Enabling Multi-Source Neural Machine Translation By Concatenating Source Sentences In Multiple Languages

TL;DR

This paper presents a simple preprocessing method for multi-source neural machine translation that concatenates source sentences in multiple languages, improving translation quality without changing the NMT architecture.

Contribution

The authors introduce a straightforward approach to MSNMT by concatenating source sentences, achieving competitive results without modifying existing NMT models.

Findings

Up to 4 BLEU improvement with 2 source languages

Up to 6 BLEU improvement with 5 source languages

Method is effective in both resource-rich and resource-poor settings

Abstract

In this paper, we explore a simple solution to "Multi-Source Neural Machine Translation" (MSNMT) which only relies on preprocessing a N-way multilingual corpus without modifying the Neural Machine Translation (NMT) architecture or training procedure. We simply concatenate the source sentences to form a single long multi-source input sentence while keeping the target side sentence as it is and train an NMT system using this preprocessed corpus. We evaluate our method in resource poor as well as resource rich settings and show its effectiveness (up to 4 BLEU using 2 source languages and up to 6 BLEU using 5 source languages). We also compare against existing methods for MSNMT and show that our solution gives competitive results despite its simplicity. We also provide some insights on how the NMT system leverages multilingual information in such a scenario by visualizing attention.