DomainLLM/gerlayqa-bgb-paraphrased

GerLayQA-BGB Paraphrased是一个针对德国民法（BGB）问题回答任务的精细调整大型语言模型特别准备的改写和重构版本的GerLayQA BGB数据集。该数据集包含5,255个高质量的关于德国民法的问题和答案对。问题已经被改写以去除抄袭，同时保持法律准确性。答案采用一致的结构分为7个部分，并包含全面的法律推理和详细解释。数据集还包括完整的法条文本，并按照90/10的比例分为训练集和验证集。问题长度不超过256个单词，答案长度不超过1024个单词，且所有内容都经过GPT-5的清洗和格式化处理。

GerLayQA-BGB Paraphrased is a paraphrased and restructured version of the GerLayQA BGB (Bürgerliches Gesetzbuch / German Civil Code) dataset, specifically prepared for fine-tuning large language models on German civil law question-answering tasks. The dataset contains 5,255 high-quality QA pairs about German Civil Law (BGB). Questions are paraphrased to remove plagiarism while maintaining legal accuracy, and answers are structured in a consistent 7-section format with comprehensive legal reasoning and detailed explanations. Full article texts are included for reference, and the dataset is split 90/10 for train/validation. The content is length-filtered and cleaned/formatted by GPT-5.