Tradutor: Building a Variety Specific Translation Model

TL;DR

This paper introduces a new open-source translation model and dataset specifically for European Portuguese, addressing the gap in language variety representation and improving translation quality over existing open-source systems.

Contribution

The paper presents the first European Portuguese-specific translation model and a dedicated dataset, enhancing translation performance for this underrepresented language variety.

Findings

Our model outperforms existing open-source Portuguese translation systems.

The model approaches industry-leading performance for European Portuguese.

Publicly available dataset and models support further research.

Abstract

Language models have become foundational to many widely used systems. However, these seemingly advantageous models are double-edged swords. While they excel in tasks related to resource-rich languages like English, they often lose the fine nuances of language forms, dialects, and varieties that are inherent to languages spoken in multiple regions of the world. Languages like European Portuguese are neglected in favor of their more popular counterpart, Brazilian Portuguese, leading to suboptimal performance in various linguistic tasks. To address this gap, we introduce the first open-source translation model specifically tailored for European Portuguese, along with a novel dataset specifically designed for this task. Results from automatic evaluations on two benchmark datasets demonstrate that our best model surpasses existing open-source translation systems for Portuguese and approaches the performance of industry-leading closed-source systems for European Portuguese. By making our dataset, models, and code publicly available, we aim to support and encourage further research, fostering advancements in the representation of underrepresented language varieties.