SIT at MixMT 2022: Fluent Translation Built on Giant Pre-trained Models

TL;DR

This paper presents a multilingual NMT system for code-mixed translation tasks, leveraging large pre-trained models, in-domain data, back-translation, and ensemble methods, achieving top rankings in WMT 2022 shared tasks.

Contribution

It introduces a high-performing translation system for Hinglish, utilizing giant pre-trained models and advanced techniques, setting new benchmarks in the shared task.

Findings

Achieved 1st place in subtask 2 (Hinglish to English) across multiple metrics.

Achieved 1st place in subtask 1 (Hindi/English to Hinglish) according to WER and human evaluation.

Secured 3rd place in subtask 1 based on ROUGE-L.

Abstract

This paper describes the Stevens Institute of Technology's submission for the WMT 2022 Shared Task: Code-mixed Machine Translation (MixMT). The task consisted of two subtasks, subtask $1$ Hindi/English to Hinglish and subtask $2$ Hinglish to English translation. Our findings lie in the improvements made through the use of large pre-trained multilingual NMT models and in-domain datasets, as well as back-translation and ensemble techniques. The translation output is automatically evaluated against the reference translations using ROUGE-L and WER. Our system achieves the $1^{st}$ position on subtask $2$ according to ROUGE-L, WER, and human evaluation, $1^{st}$ position on subtask $1$ according to WER and human evaluation, and $3^{rd}$ position on subtask $1$ with respect to ROUGE-L metric.