Toward Multilingual Neural Machine Translation with Universal Encoder and Decoder

TL;DR

This paper introduces a unified multilingual neural machine translation framework using attention-based models, capable of handling many-to-many translation tasks efficiently without special architecture modifications.

Contribution

It presents a universal encoder-decoder approach that learns minimal parameters and performs well in low-resource and zero-resource translation scenarios.

Findings

Achieved up to 2.6 BLEU points improvement in under-resourced translation tasks.

Demonstrated effectiveness in zero-resource translation without direct parallel corpora.

Utilized standard training methods without special architectural modifications.

Abstract

In this paper, we present our first attempts in building a multilingual Neural Machine Translation framework under a unified approach. We are then able to employ attention-based NMT for many-to-many multilingual translation tasks. Our approach does not require any special treatment on the network architecture and it allows us to learn minimal number of free parameters in a standard way of training. Our approach has shown its effectiveness in an under-resourced translation scenario with considerable improvements up to 2.6 BLEU points. In addition, the approach has achieved interesting and promising results when applied in the translation task that there is no direct parallel corpus between source and target languages.