Efficiently Exploring Large Language Models for Document-Level Machine Translation with In-context Learning

TL;DR

This paper introduces CAP, a context-aware prompting method that improves the coherence and accuracy of document-level machine translation using large language models by selecting relevant context and summaries for better in-context learning.

Contribution

The paper proposes a novel context-aware prompting technique that enhances LLMs' performance in document translation by selecting relevant sentences and generating summaries for improved coherence.

Findings

CAP improves translation coherence and accuracy

Effective in zero pronoun and literary translation tasks

Demonstrates significant gains across various DOCMT benchmarks

Abstract

Large language models (LLMs) exhibit outstanding performance in machine translation via in-context learning. In contrast to sentence-level translation, document-level translation (DOCMT) by LLMs based on in-context learning faces two major challenges: firstly, document translations generated by LLMs are often incoherent; secondly, the length of demonstration for in-context learning is usually limited. To address these issues, we propose a Context-Aware Prompting method (CAP), which enables LLMs to generate more accurate, cohesive, and coherent translations via in-context learning. CAP takes into account multi-level attention, selects the most relevant sentences to the current one as context, and then generates a summary from these collected sentences. Subsequently, sentences most similar to the summary are retrieved from the datastore as demonstrations, which effectively guide LLMs in generating cohesive and coherent translations. We conduct extensive experiments across various DOCMT tasks, and the results demonstrate the effectiveness of our approach, particularly in zero pronoun translation (ZPT) and literary translation tasks.