Machine Translation with Large Language Models: Decoder Only vs. Encoder-Decoder

TL;DR

This paper compares decoder-only and encoder-decoder large language models for multilingual machine translation, focusing on Indian languages, to identify which architecture offers better translation quality and efficiency.

Contribution

It provides a comparative analysis of decoder-only and encoder-decoder architectures for multilingual translation, especially for Indian languages, highlighting their relative strengths and weaknesses.

Findings

Encoder-decoder models outperform decoder-only models in translation accuracy.

Decoder-only models are more efficient but less accurate.

The study offers insights into architecture selection for multilingual MT.

Abstract

This project, titled "Machine Translation with Large Language Models: Decoder-only vs. Encoder-Decoder," aims to develop a multilingual machine translation (MT) model. Focused on Indian regional languages, especially Telugu, Tamil, and Malayalam, the model seeks to enable accurate and contextually appropriate translations across diverse language pairs. By comparing Decoder-only and Encoder-Decoder architectures, the project aims to optimize translation quality and efficiency, advancing cross-linguistic communication tools.The primary objective is to develop a model capable of delivering high-quality translations that are accurate and contextually appropriate. By leveraging large language models, specifically comparing the effectiveness of Decoder-only and Encoder-Decoder architectures, the project seeks to optimize translation performance and efficiency across multilingual contexts. Through rigorous experimentation and analysis, this project aims to advance the field of machine translation, contributing valuable insights into the effectiveness of different model architectures and paving the way for enhanced cross-linguistic communication tools.