Controlling Translation Formality Using Pre-trained Multilingual Language Models

TL;DR

This paper explores using a single pre-trained multilingual model to control translation formality across six languages, achieving comparable quality to dedicated models but heavily influenced by model and data choices.

Contribution

It demonstrates the feasibility of formality control in multilingual translation with a single model, highlighting the importance of pre-trained model selection and finetuning data.

Findings

Single multilingual model can control formality effectively

Model choice and finetuning data significantly impact results

Approaches can match dedicated translation model quality

Abstract

This paper describes the University of Maryland's submission to the Special Task on Formality Control for Spoken Language Translation at \iwslt, which evaluates translation from English into 6 languages with diverse grammatical formality markers. We investigate to what extent this problem can be addressed with a \textit{single multilingual model}, simultaneously controlling its output for target language and formality. Results show that this strategy can approach the translation quality and formality control achieved by dedicated translation models. However, the nature of the underlying pre-trained language model and of the finetuning samples greatly impact results.