翻译的时机与位置？翻译策略对跨语言大语言模型提示的影响
How and Where to Translate? The Impact of Translation Strategies in Cross-lingual LLM Prompting
AI总结：
该研究针对跨语言RAG系统中知识库与上下文语言不一致的问题，系统评估不同提示翻译策略对RAG增强LLM分类任务的影响，发现优化策略可显著提升跨语言知识共享与下游性能，为低资源语言场景提供优化方向。
摘要：
尽管大语言模型（LLM）的多语言能力不断进步，其性能在不同语言和任务间仍存在显著差异。在基于多语言检索增强生成（RAG）的系统中，知识库（KB）通常从高资源语言（如英语）共享到低资源语言，导致从知识库检索到的信息与上下文其余部分的语言不同。在这类场景下，两种常见做法是预翻译以构建单语言提示，以及采用跨语言提示直接推理。然而，这些选择的影响仍不明确。本文系统评估了不同提示翻译策略在多语言系统中RAG增强LLM的分类任务上的影响。实验结果表明，优化的提示策略能显著提升跨语言知识共享效果，进而提升下游分类任务的性能。研究结果主张在非英语语言（尤其是低资源语言）场景中更广泛地应用多语言资源共享和跨语言提示优化。
英文摘要：
Despite advances in the multilingual capabilities of Large Language Models (LLMs), their performance varies substantially across different languages and tasks. In multilingual retrieval-augmented generation (RAG)-based systems, knowledge bases (KB) are often shared from high-resource languages (such as English) to low-resource ones, resulting in retrieved information from the KB being in a different language than the rest of the context. In such scenarios, two common practices are pre-translation to create a mono-lingual prompt and cross-lingual prompting for direct inference. However, the impact of these choices remains unclear. In this paper, we systematically evaluate the impact of different prompt translation strategies for classification tasks with RAG-enhanced LLMs in multilingual systems. Experimental results show that an optimized prompting strategy can significantly improve knowledge sharing across languages, therefore improve the performance on the downstream classification task. The findings advocate for a broader utilization of multilingual resource sharing and cross-lingual prompt optimization for non-English languages, especially the low-resource ones.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
