marvintong/legal-llm-benchmark
收藏官方服务：
资源简介：
Legal LLM Benchmark数据集用于评估12个大语言模型在163个法律任务上的安全性和实用性权衡。它包含多个子数据集，如问题、评估、合同分析、法律实践领域分类等，用于研究和教育目的。
The Legal LLM Benchmark dataset is designed to evaluate the safety-utility trade-off of 12 large language models across 163 legal tasks. It includes multiple sub-datasets such as questions, evaluations, contract analysis, legal practice area taxonomy, etc., for research and educational purposes.
提供机构：
marvintong
