用多轮对话增强翻译连贯性，无需训练即可提升文档级翻译质量。
Source-primed Multi-turn Conversation Helps Large Language Models Translate Documents
- 分段迭代翻译并保留上下文，实现无训练连贯译文。
- 相比单次翻译或独立分段翻译，自动评测指标全面领先。
- 复用历史键值缓存，计算开销极低，适合实际部署。
大语言模型已推动文档级机器翻译的简化，但遗漏错误等问题仍存在。本文研究一种通过多轮对话方式利用前序上下文处理文档级翻译的简单方法。具体而言，将文档分解为多个段落，逐段进行迭代翻译并保持历史对话记录，该方法在无需额外训练的前提下确保翻译连贯性，并可完全复用先前对话的键值缓存（KV cache），显著降低计算开销。我们进一步提出“源端驱动”策略，在多轮翻译前先提供完整原文。实验证明，该方法在多个主流大模型上均优于单次整体翻译和独立分段翻译，在多项自动评测指标上表现更优，为基于大语言模型的文档级翻译建立了强基准。
原文摘要 · Abstract (English)
LLMs have paved the way for truly simple document-level machine translation, but challenges such as omission errors remain. In this paper, we study a simple method for handling document-level machine translation, by leveraging previous contexts in a multi-turn conversational manner. Specifically, by decomposing documents into segments and iteratively translating them while maintaining previous turns, this method ensures coherent translations without additional training, and can fully re-use the KV cache of previous turns thus minimizing computational overhead. We further propose a `source-primed' method that first provides the whole source document before multi-turn translation. We empirically show this multi-turn method outperforms both translating entire documents in a single turn and translating each segment independently according to multiple automatic metrics in representative LLMs, establishing a strong baseline for document-level translation using LLMs.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。