LLM-based Translation Inference with Iterative Bilingual Understanding

TL;DR

This paper introduces IBUT, an iterative translation method leveraging LLMs' cross-lingual understanding to refine translations through feedback, significantly improving performance across multiple domains.

Contribution

The paper presents a novel IBUT approach that uses iterative bilingual understanding with LLMs to enhance translation accuracy, addressing errors from initial misunderstandings.

Findings

IBUT outperforms several strong comparison methods.

IBUT generalizes well across multiple domains.

Experimental results show improved translation quality.

Abstract

The remarkable understanding and generation capabilities of large language models (LLMs) have greatly improved translation performance. However, incorrect understanding of the sentence to be translated can degrade translation quality. To address this issue, we proposed a novel Iterative Bilingual Understanding Translation (IBUT) method based on the cross-lingual capabilities of LLMs and the dual characteristics of translation tasks. The cross-lingual capability of LLMs enables the generation of contextual understanding for both the source and target languages separately. Furthermore, the dual characteristics allow IBUT to generate effective cross-lingual feedback, iteratively refining contextual understanding, thereby reducing errors and improving translation performance. Experimental results showed that the proposed IBUT outperforms several strong comparison methods, especially being generalized to multiple domains (e.g., news, commonsense, and cultural translation benchmarks).