用大模型生成符合人类偏好的多语言翻译数据，让机器翻译更自然。
PMMT: Preference Alignment in Multilingual Machine Translation via LLM Distillation
- 用大模型生成带特定风格偏好（如正式/口语）的多语言语料
- 在未训练过的WMT和Flores数据集上表现媲美当前最优方法
- 适合需要低成本、高效率部署个性化翻译服务的场景
翻译对跨语言交流至关重要，尽管已有大量工作提升翻译准确性，但针对人类偏好（如语气、风格）对齐的研究仍不足。本文提出一种新方法，利用大语言模型（LLM）高效生成大规模带特定翻译偏好的多语言平行语料。同时设计自动化流水线，将人类偏好蒸馏至小型机器翻译（MT）模型中，实现在线服务的大规模、低成本调用。实验表明，该方法在具有人类偏好对齐任务中大幅领先；在未参与训练的主流公开基准（如WMT、Flores）上，性能也与当前最优方法相当。
原文摘要 · Abstract (English)
Translation is important for cross-language communication, and many efforts have been made to improve its accuracy. However, less investment is conducted in aligning translations with human preferences, such as translation tones or styles. In this paper, a new method is proposed to effectively generate large-scale multilingual parallel corpora with specific translation preferences using Large Language Models (LLMs). Meanwhile, an automatic pipeline is designed to distill human preferences into smaller Machine Translation (MT) models for efficiently and economically supporting large-scale calls in online services. Experiments indicate that the proposed method takes the lead in translation tasks with aligned human preferences by a large margin. Meanwhile, on popular public benchmarks like WMT and Flores, on which our models were not trained, the proposed method also shows a competitive performance compared to SOTA works.
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。