LLMs Are Zero-Shot Context-Aware Simultaneous Translators

Roman Koshkin, Katsuhito Sudoh, Satoshi Nakamura
6/19/2024
cs.CL

No official code found. Best implementation: RomanKoshkin/toLLMatch (10 stars, MIT, last commit 2025-03-06).

Abstract

The advent of transformers has fueled progress in machine translation. More recently large language models (LLMs) have come to the spotlight thanks to their generality and strong performance in a wide range of language tasks, including translation. Here we show that open-source LLMs perform on par with or better than some state-of-the-art baselines in simultaneous machine translation (SiMT) tasks, zero-shot. We also demonstrate that injection of minimal background information, which is easy with an LLM, brings further performance gains, especially on challenging technical subject-matter. This highlights LLMs' potential for building next generation of massively multilingual, context-aware and terminologically accurate SiMT systems that require no resource-intensive training or fine-tuning.

View on arXivView PDF

Code Implementations(1)

RomanKoshkin/toLLMatchCommunity100%
MIT

toLLMatch🔪: Context-aware LLM-based simultaneous translation

103C, C++Jun 20, 20241 years agoMIT

Similar Papers

TransLLaMa: LLM-based Simultaneous Translation System

91% match2024

LLMs Can Achieve High-quality Simultaneous Machine Translation as Efficiently as Offline

87% match1 repo82025

Beyond the Sentence: A Survey on Context-Aware Machine Translation with Large Language Models

86% match2025

Towards Boosting Many-to-Many Multilingual Machine Translation with Large Language Models

86% match1 repo22024

SiLLM: Large Language Models for Simultaneous Machine Translation

86% match1 repo182024

Compensating for Data with Reasoning: Low-Resource Machine Translation with LLMs

86% match2025

Cite this paper

@article{koshkin2024llms,
  title  = {LLMs Are Zero-Shot Context-Aware Simultaneous Translators},
  author = {Roman Koshkin and Katsuhito Sudoh and Satoshi Nakamura},
  year   = {2024},
  eprint = {2406.13476},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2406.13476v3}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2406.13476.svg)](https://codeofpaper.com/paper/2406.13476)

Discussion