We propose the on-the-fly ensembling of a machine translation model with an LLM, prompted on the same task and input. We perform experiments on 4 language pairs (both directions) with varying data amounts. We find that a slightly weaker-at-translation LLM can improve translations of a NMT model, and ensembling with an LLM can produce better translations than ensembling two stronger MT models. We combine our method with various techniques from LLM prompting, such as in context learning and translation context.

翻译：我们提出了一种将机器翻译模型与大语言模型（LLM）进行即时集成的方法，该方法基于相同任务和输入的提示。我们在4个语言对（双向）上进行了实验，涉及不同数据量。研究发现，在翻译能力上稍弱的LLM能够提升神经机器翻译（NMT）模型的翻译质量，而将LLM与NMT模型集成所获得的翻译效果优于集成两个更强的机器翻译模型。我们将该方法与多种LLM提示技术（如上下文学习和翻译上下文）相结合。