D-SCoRE：基于文档中心分割与推理的结构化导出QA-CoT数据生成
D-SCoRE: Document-Centric Segmentation and CoT Reasoning with Structured Export for QA-CoT Data Generation
AI总结：
D-SCoRE通过文档中心分割与推理生成高质量QA-CoT数据，提升领域适应微调性能。
摘要：
高质量领域特定问答（QA）数据的稀缺性和高成本限制了大型语言模型（LLMs）的监督微调。我们引入D-SCoRE，一种无需训练的框架，利用LLMs和提示工程自动从任意文本源生成多样、丰富的QA数据集，通过链式推理（CoT）。通过整合文档中心处理、分割、推理推理和结构化导出，以及语义角色转换、问题类型平衡和反事实增强等多维控制，D-SCoRE生成定制化的QA对，具有增强的多样性和相关性。在大多数评估领域中，在D-SCoRE生成的数据集上微调的LLMs在人类标注的QA数据上表现更优。其效率和可扩展性使在消费级硬件上实现快速、高性能的领域适应微调成为可能，每GPU小时可生成超过1,100对高质量QA对。
英文摘要：
The scarcity and high cost of high-quality domain-specific question-answering (QA) datasets limit supervised fine-tuning of large language models (LLMs). We introduce $\textbf{D-SCoRE}$, a training-free framework that leverages LLMs and prompt engineering to automatically generate diverse, rich QA datasets with Chain-of-Thought (CoT) from arbitrary textual sources. By integrating $\textbf{D}$ocument-centric processing, $\textbf{S}$egmentation, $\textbf{Co}$T $\textbf{R}$easoning, and structured $\textbf{E}$xport - along with multi-dimensional controls such as semantic role transformation, question type balancing, and counterfactual augmentation - D-SCoRE produces tailored QA pairs with enhanced diversity and relevance. LLMs fine-tuned on D-SCoRE-generated datasets outperform those trained on human-annotated QA data across most evaluated domains. Its efficiency and scalability enable rapid, high-performance domain-adaptive fine-tuning on consumer-grade hardware, generating over 1,100 high-quality QA pairs per GPU-hour end-to-end.
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。
