sPhinX：通过N-shot引导提示实现样本高效的多语言指令微调
sPhinX: Sample Efficient Multilingual Instruction Fine-Tuning Through N-shot Guided Prompting
AI总结：
本文提出sPhinX多语言指令微调数据集构建方法及LANGIT的N-shot引导策略，显著提升Mistral-7B和Phi-3-Small在多语言基准上的性能，同时保持英语能力且无灾难性遗忘。
摘要：
尽管大型语言模型（LLMs）在英语方面取得了显著成功，但在非英语语言中仍存在显著的性能差距。为解决这一问题，我们提出了一种新颖的方法，用于策略性地构建多语言合成指令微调数据集sPhinX。与先前直接翻译固定指令-响应对的方法不同，sPhinX通过选择性地用多语言翻译增强英语指令-响应对来提升多样性。此外，我们提出了LANGIT，一种新颖的N-shot引导微调策略，该策略通过在每个训练样本中加入上下文相关的示例来进一步提升模型性能。我们的消融研究表明，我们的方法增强了Mistral-7B和Phi-3-Small的多语言能力，在多语言基准测试中，推理、问答、阅读理解及机器翻译方面的平均性能分别提升了39.8%和11.2%。此外，即使训练涉及51种语言，sPhinX在英语LLM基准测试中仍保持强劲性能，且表现出极小甚至无灾难性遗忘。
英文摘要：
Despite the remarkable success of large language models (LLMs) in English, a significant performance gap remains in non-English languages. To address this, we introduce a novel approach for strategically constructing a multilingual synthetic instruction tuning dataset, sPhinX. Unlike prior methods that directly translate fixed instruction-response pairs, sPhinX enhances diversity by selectively augmenting English instruction-response pairs with multilingual translations. Additionally, we propose LANGIT, a novel N-shot guided fine-tuning strategy, which further enhances model performance by incorporating contextually relevant examples in each training sample. Our ablation study shows that our approach enhances the multilingual capabilities of Mistral-7B and Phi-3-Small improving performance by an average of 39.8% and 11.2%, respectively, across multilingual benchmarks in reasoning, question answering, reading comprehension, and machine translation. Moreover, sPhinX maintains strong performance on English LLM benchmarks while exhibiting minimal to no catastrophic forgetting, even when trained on 51 languages.
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。
