Consistency is a key requirement of high-quality translation. It is especially important to adhere to pre-approved terminology and adapt to corrected translations in domain-specific projects. Machine translation (MT) has achieved significant progress in the area of domain adaptation. However, real-time adaptation remains challenging. Large-scale language models (LLMs) have recently shown interesting capabilities of in-context learning, where they learn to replicate certain input-output text generation patterns, without further fine-tuning. By feeding an LLM at inference time with a prompt that consists of a list of translation pairs, it can then simulate the domain and style characteristics. This work aims to investigate how we can utilize in-context learning to improve real-time adaptive MT. Our extensive experiments show promising results at translation time. For example, GPT-3.5 can adapt to a set of in-domain sentence pairs and/or terminology while translating a new sentence. We observe that the translation quality with few-shot in-context learning can surpass that of strong encoder-decoder MT systems, especially for high-resource languages. Moreover, we investigate whether we can combine MT from strong encoder-decoder models with fuzzy matches, which can further improve translation quality, especially for less supported languages. We conduct our experiments across five diverse language pairs, namely English-to-Arabic (EN-AR), English-to-Chinese (EN-ZH), English-to-French (EN-FR), English-to-Kinyarwanda (EN-RW), and English-to-Spanish (EN-ES).

翻译：一致性是高质量翻译的关键要求。在特定领域的项目中，遵循预先批准的术语并适应修正后的翻译尤为重要。机器翻译（MT）在领域自适应方面已取得显著进展，但实时自适应仍具挑战性。大语言模型（LLM）近期展现出有趣的上下文学习能力——无需进一步微调，即可通过模仿特定输入-输出文本生成模式进行学习。在推理阶段，通过向LLM提供包含翻译对列表的提示词，模型能够模拟目标领域和风格特征。本研究旨在探索如何利用上下文学习改进实时自适应机器翻译。我们的大量实验在翻译时展现出可喜成果。例如，GPT-3.5在翻译新句子时可自适应一组领域内句子对和/或术语。我们观察到，基于少样本上下文学习的翻译质量可超越强编码器-解码器MT系统，尤其对于高资源语言。此外，我们研究了能否将强编码器-解码器模型的MT能力与模糊匹配相结合，从而进一步提升翻译质量，尤其是对资源较少的语言支持。实验涵盖五种不同语言对：英语-阿拉伯语（EN-AR）、英语-中文（EN-ZH）、英语-法语（EN-FR）、英语-卢旺达语（EN-RW）及英语-西班牙语（EN-ES）。