NLIP_Lab-IITH Multilingual MT System for WAT24 MT Shared Task

TL;DR

This paper presents a multilingual machine translation system for 22 Indic languages, utilizing pre-training, bilingual dictionaries, and fine-tuning, achieving competitive results on multiple benchmarks.

Contribution

Introduces a 243M parameter multilingual translation model for 22 Indic languages with novel pre-training and dictionary-based techniques, advancing multilingual Indic MT performance.

Findings

Achieved high chrF++ and BLEU scores on WAT24 benchmarks.

Model is competitive with larger IndicTransv1 system.

Effective use of bilingual dictionaries and fine-tuning improves translation quality.

Abstract

This paper describes NLIP Lab's multilingual machine translation system for the WAT24 shared task on multilingual Indic MT task for 22 scheduled languages belonging to 4 language families. We explore pre-training for Indic languages using alignment agreement objectives. We utilize bi-lingual dictionaries to substitute words from source sentences. Furthermore, we fine-tuned language direction-specific multilingual translation models using small and high-quality seed data. Our primary submission is a 243M parameters multilingual translation model covering 22 Indic languages. In the IN22-Gen benchmark, we achieved an average chrF++ score of 46.80 and 18.19 BLEU score for the En-Indic direction. In the Indic-En direction, we achieved an average chrF++ score of 56.34 and 30.82 BLEU score. In the In22-Conv benchmark, we achieved an average chrF++ score of 43.43 and BLEU score of 16.58 in the En-Indic direction, and in the Indic-En direction, we achieved an average of 52.44 and 29.77 for chrF++ and BLEU respectively. Our model\footnote{Our code and models are available at \url{https://github.com/maharajbrahma/WAT2024-MultiIndicMT}} is competitive with IndicTransv1 (474M parameter model).