nyu-dice-lab/sos-artifacts
收藏Hugging Face2025-05-21 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/nyu-dice-lab/sos-artifacts
下载链接
链接失效反馈官方服务：
资源简介：
Arena-Hard-Auto基准数据集，包含不同模型的回答和评判，用于研究大型语言模型在基准测试中的表现。数据集还包括排行榜、消融研究以及评判AI回答的详细标准。
The Arena-Hard-Auto benchmark dataset, which includes responses and judgments from various models, is used to study the performance of large language models in benchmark tests. The dataset also includes leaderboards, ablation studies, and detailed criteria for judging AI responses.
提供机构：
nyu-dice-lab
