LexNum
收藏资源简介:
LexNum是首个专门为中文法律场景中的数学推理任务设计的法律数据集,由中国人民大学高薪人工智能学院和国际商务与经济大学构建。该数据集覆盖了三种高频诉讼场景:经济补偿、工伤赔偿和交通事故赔偿。数据来源于pkulaw网站,通过LLM模型提取与法律数学推理相关的内容,并经过人类专家的双重验证以保证数据质量。该数据集旨在为法律场景中的LLM推理研究提供支持。
LexNum is the first specialized legal dataset designed for mathematical reasoning tasks in Chinese legal scenarios, jointly constructed by the Gaoling School of Artificial Intelligence at Renmin University of China and the University of International Business and Economics. The dataset covers three high-frequency litigation scenarios: economic compensation, work-related injury compensation, and traffic accident compensation. Its data is sourced from the PKULAW website, where content related to legal mathematical reasoning is extracted via large language models (LLMs), and the data quality is ensured through double verification by human experts. This dataset aims to support research on LLM-based reasoning in legal scenarios.




