Simultaneous Machine Translation with Large Language Models

TL;DR

This paper explores using Large Language Models for simultaneous machine translation, demonstrating their advantages in quality, robustness, and flexibility, while addressing latency with a new algorithm, despite high computational costs.

Contribution

It introduces a novel RALCP algorithm for latency reduction and applies LLMs to SimulMT, showing improved performance over dedicated models in multiple languages.

Findings

LLMs outperform dedicated MT models in BLEU and LAAL metrics

LLMs show better robustness and tuning efficiency

Computational cost remains a significant challenge

Abstract

Real-world simultaneous machine translation (SimulMT) systems face more challenges than just the quality-latency trade-off. They also need to address issues related to robustness with noisy input, processing long contexts, and flexibility for knowledge injection. These challenges demand models with strong language understanding and generation capabilities which may not often equipped by dedicated MT models. In this paper, we investigate the possibility of applying Large Language Models (LLM) to SimulMT tasks by using existing incremental-decoding methods with a newly proposed RALCP algorithm for latency reduction. We conducted experiments using the \texttt{Llama2-7b-chat} model on nine different languages from the MUST-C dataset. The results show that LLM outperforms dedicated MT models in terms of BLEU and LAAL metrics. Further analysis indicates that LLM has advantages in terms of tuning efficiency and robustness. However, it is important to note that the computational cost of LLM remains a significant obstacle to its application in SimulMT.