johnsonjp34/Due-Process-Training
收藏Hugging Face2025-12-18 更新2025-12-20 收录
下载链接：
https://hf-mirror.com/datasets/johnsonjp34/Due-Process-Training
下载链接
链接失效反馈官方服务：
资源简介：
该数据集用于通过LoRA方法微调语言模型，以识别美国法院案件中的正当程序问题。数据集包含5,000个经过清理和注释的法院案例，每个案例都包含清理后的法院意见文本、相关案件事实的摘要以及正当程序问题的分析。数据以JSON Lines格式存储，适用于基于指令的监督微调。
This dataset is used to fine-tune a language model (via LoRA) to identify due process issues in U.S. court cases. The dataset consists of 5,000 court cases that have been cleaned and annotated, each containing cleaned court opinion text, a summary of relevant case facts, and an analysis of due process issues. The data is stored in JSON Lines format and is suitable for instruction-based supervised fine-tuning.
提供机构：
johnsonjp34
