Together We Can: Multilingual Automatic Post-Editing for Low-Resource Languages

TL;DR

This study explores multilingual Automatic Post-Editing (APE) for low-resource Indo-Aryan languages, leveraging linguistic similarities and synthetic data to improve translation quality through multi-task learning and domain adaptation.

Contribution

It introduces a multilingual APE model for Hindi and Marathi, utilizing synthetic triplets and multi-task learning to enhance translation quality for low-resource languages.

Findings

Multilingual APE outperforms single-pair models by 2.5 and 2.39 TER points.

Multi-task learning further improves performance by 1.29 and 1.44 TER points.

Data augmentation and domain adaptation yield additional TER improvements.

Abstract

This exploratory study investigates the potential of multilingual Automatic Post-Editing (APE) systems to enhance the quality of machine translations for low-resource Indo-Aryan languages. Focusing on two closely related language pairs, English-Marathi and English-Hindi, we exploit the linguistic similarities to develop a robust multilingual APE model. To facilitate cross-linguistic transfer, we generate synthetic Hindi-Marathi and Marathi-Hindi APE triplets. Additionally, we incorporate a Quality Estimation (QE)-APE multi-task learning framework. While the experimental results underline the complementary nature of APE and QE, we also observe that QE-APE multitask learning facilitates effective domain adaptation. Our experiments demonstrate that the multilingual APE models outperform their corresponding English-Hindi and English-Marathi single-pair models by $2.5$ and $2.39$ TER points, respectively, with further notable improvements over the multilingual APE model observed through multi-task learning ($+1.29$ and $+1.44$ TER points), data augmentation ($+0.53$ and $+0.45$ TER points) and domain adaptation ($+0.35$ and $+0.45$ TER points). We release the synthetic data, code, and models accrued during this study publicly at https://github.com/cfiltnlp/Multilingual-APE.