UvA-MT's Participation in the WMT23 General Translation Shared Task

TL;DR

This paper presents UvA-MT's approach to the WMT23 shared task, demonstrating that a single multilingual model with specific strategies can achieve competitive results in English-Hebrew translation.

Contribution

It introduces a minimal multilingual model with strategies like back-translation and fine-tuning that performs comparably to bilingual models in translation tasks.

Findings

Bidirectional single models can match bilingual performance.

Effective strategies improve translation quality.

Competitive results achieved in automatic evaluation.

Abstract

This paper describes the UvA-MT's submission to the WMT 2023 shared task on general machine translation. We participate in the constrained track in two directions: English <-> Hebrew. In this competition, we show that by using one model to handle bidirectional tasks, as a minimal setting of Multilingual Machine Translation (MMT), it is possible to achieve comparable results with that of traditional bilingual translation for both directions. By including effective strategies, like back-translation, re-parameterized embedding table, and task-oriented fine-tuning, we obtained competitive final results in the automatic evaluation for both English -> Hebrew and Hebrew -> English directions.