TEaR: Improving LLM-based Machine Translation with Systematic Self-Refinement

TL;DR

TEaR introduces a systematic self-refinement framework for LLM-based machine translation, enabling iterative error correction that improves translation quality across diverse languages with enhanced interpretability and strategic estimation methods.

Contribution

The paper presents TEaR, a novel self-refinement framework for LLMs that systematically improves translation accuracy through error estimation and iterative correction, demonstrating broad language applicability.

Findings

TEaR improves translation quality across multiple languages.

The framework enhances systematicity and interpretability.

Different estimation strategies impact correction effectiveness.

Abstract

Large Language Models (LLMs) have achieved impressive results in Machine Translation (MT). However, careful evaluations by human reveal that the translations produced by LLMs still contain multiple errors. Importantly, feeding back such error information into the LLMs can lead to self-refinement and result in improved translation performance. Motivated by these insights, we introduce a systematic LLM-based self-refinement translation framework, named \textbf{TEaR}, which stands for \textbf{T}ranslate, \textbf{E}stimate, \textbf{a}nd \textbf{R}efine, marking a significant step forward in this direction. Our findings demonstrate that 1) our self-refinement framework successfully assists LLMs in improving their translation quality across a wide range of languages, whether it's from high-resource languages to low-resource ones or whether it's English-centric or centered around other languages; 2) TEaR exhibits superior systematicity and interpretability; 3) different estimation strategies yield varied impacts, directly affecting the effectiveness of the final corrections. Additionally, traditional neural translation models and evaluation models operate separately, often focusing on singular tasks due to their limited capabilities, while general-purpose LLMs possess the capability to undertake both tasks simultaneously. We further conduct cross-model correction experiments to investigate the potential relationship between the translation and evaluation capabilities of general-purpose LLMs. Our code and data are available at https://github.com/fzp0424/self_correct_mt