arXiv 2401.070372024-01-13cs.CLcs.AI

xCoT：面向跨语言思维链推理的跨语言指令微调

xCoT: Cross-lingual Instruction Tuning for Cross-lingual Chain-of-Thought Reasoning

State Key Lab of Software Development Environment, Beihang University（北京航空航天大学软件开发环境国家重点实验室）；Beijing Information Science and Technology University（北京信息科技大学）；School of New Media and Communication, Tianjin University（天津大学新媒体与传播学院）

Linzheng Chai, Jian Yang, Tao Sun, Hongcheng Guo, Jiaheng Liu, Bing Wang, Xiannian Liang, Jiaqi Bai, Tongliang Li, Qiyao Peng, Zhoujun Li

备注：: 11 pages

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出跨语言指令微调框架xCoT，通过多语言指令数据、跨语言少样本学习、随机在线CoT和跨语言蒸馏，将高资源语言的思维链推理能力迁移到低资源语言，显著缩小跨语言性能差距。

摘要：

思维链（CoT）已成为一种强大的技术，用于激发大型语言模型中的推理能力并改进多种下游任务。CoT主要在英语中表现出卓越性能，但由于语言泛化能力较差，其在低资源语言中的使用受到限制。为了弥合不同语言之间的差距，我们提出了一种跨语言指令微调框架（xCOT），将知识从高资源语言迁移到低资源语言。具体来说，我们创建了多语言指令训练数据（xCOT-INSTRUCT），以促进多种语言的语义对齐。我们引入跨语言上下文少样本学习（xICL）来加速指令微调中的多语言一致性，其中示例中源语言的某些片段被随机替换为目标语言的对应翻译。在多语言指令微调过程中，我们采用随机在线CoT策略，通过先将查询翻译成另一种语言再用英语回答，来增强大型语言模型的多语言推理能力。为了进一步促进语言迁移，我们利用高资源CoT通过跨语言蒸馏来监督低资源语言的训练。在以往基准上的实验结果表明，xCoT在缩小不同语言之间差距方面表现出优越性能，凸显了其缩小跨语言差距的潜力。

英文摘要：

Chain-of-thought (CoT) has emerged as a powerful technique to elicit reasoning in large language models and improve a variety of downstream tasks. CoT mainly demonstrates excellent performance in English, but its usage in low-resource languages is constrained due to poor language generalization. To bridge the gap among different languages, we propose a cross-lingual instruction fine-tuning framework (xCOT) to transfer knowledge from high-resource languages to low-resource languages. Specifically, the multilingual instruction training data (xCOT-INSTRUCT) is created to encourage the semantic alignment of multiple languages. We introduce cross-lingual in-context few-shot learning (xICL)) to accelerate multilingual agreement in instruction tuning, where some fragments of source languages in examples are randomly substituted by their counterpart translations of target languages. During multilingual instruction tuning, we adopt the randomly online CoT strategy to enhance the multilingual reasoning ability of the large language model by first translating the query to another language and then answering in English. To further facilitate the language transfer, we leverage the high-resource CoT to supervise the training of low-resource languages with cross-lingual distillation. Experimental results on previous benchmarks demonstrate the superior performance of xCoT in reducing the gap among different languages, highlighting its potential to reduce the cross-lingual gap.

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。