Controlling Neural Machine Translation Formality with Synthetic Supervision

TL;DR

This paper presents a neural method for controlling the formality level in machine translation by generating synthetic training data, leading to translations that better match desired formality without losing meaning.

Contribution

It introduces a novel multi-task training scheme that automatically creates synthetic triplets, enabling end-to-end training for formality control in neural machine translation.

Findings

Outperforms existing models in formality control accuracy

Produces translations that better match target formality levels

Preserves source meaning effectively

Abstract

This work aims to produce translations that convey source language content at a formality level that is appropriate for a particular audience. Framing this problem as a neural sequence-to-sequence task ideally requires training triplets consisting of a bilingual sentence pair labeled with target language formality. However, in practice, available training examples are limited to English sentence pairs of different styles, and bilingual parallel sentences of unknown formality. We introduce a novel training scheme for multi-task models that automatically generates synthetic training triplets by inferring the missing element on the fly, thus enabling end-to-end training. Comprehensive automatic and human assessments show that our best model outperforms existing models by producing translations that better match desired formality levels while preserving the source meaning.