relai-ai/legal-scenarios-SCOTUS-2024-decisions

该数据集是一个用于评估大型语言模型在法律背景下推理能力的评估集。它包含了一系列基于2024年美国最高法院意见书中的真实场景问题，要求模型提供基于先例的正确法律解决方案并进行充分论证。数据集通过RELAI的数据代理创建，可用于模型性能评估或后期训练。

This dataset is an evaluation set for assessing the reasoning ability of large language models in a legal context. It contains a series of realistic scenarios based on the slip opinions of the US Supreme Court from the 2024 term, requiring the model to provide a correct legal resolution with sufficient justification based on precedent. The dataset is created using RELAIs data agent and can be used for model performance evaluation or post-training.