Segment-Based Interactive Machine Translation for Pre-trained Models

TL;DR

This paper investigates the use of pre-trained multilingual models like mBART and mT5 in interactive machine translation, demonstrating that these models can achieve state-of-the-art performance with user feedback in translation tasks.

Contribution

It introduces a novel approach of applying pre-trained large language models to interactive machine translation, comparing their performance with existing models on benchmark datasets.

Findings

mBART performs comparably with state-of-the-art models in IMT.

Pre-trained models can effectively generate perfect translations with user feedback.

The study highlights the potential of adapting pre-trained models for interactive translation environments.

Abstract

Pre-trained large language models (LLM) are starting to be widely used in many applications. In this work, we explore the use of these models in interactive machine translation (IMT) environments. In particular, we have chosen mBART (multilingual Bidirectional and Auto-Regressive Transformer) and mT5 (multilingual Text-to-Text Transfer Transformer) as the LLMs to perform our experiments. The system generates perfect translations interactively using the feedback provided by the user at each iteration. The Neural Machine Translation (NMT) model generates a preliminary hypothesis with the feedback, and the user validates new correct segments and performs a word correction--repeating the process until the sentence is correctly translated. We compared the performance of mBART, mT5, and a state-of-the-art (SoTA) machine translation model on a benchmark dataset regarding user effort, Word Stroke Ratio (WSR), Key Stroke Ratio (KSR), and Mouse Action Ratio (MAR). The experimental results indicate that mBART performed comparably with SoTA models, suggesting that it is a viable option for this field of IMT. The implications of this finding extend to the development of new machine translation models for interactive environments, as it indicates that some novel pre-trained models exhibit SoTA performance in this domain, highlighting the potential benefits of adapting these models to specific needs.