Salesforce/SIMPLE
收藏Hugging Face2025-02-24 更新2025-04-08 收录
下载链接：
https://hf-mirror.com/datasets/Salesforce/SIMPLE
下载链接
链接失效反馈官方服务：
资源简介：
SIMPLE是一个简单推理能力测试的基准，旨在评估AI模型在类似于高中生能够手工解决的问题上的表现。当前版本包含225个问题的初步子集。
SIMPLE is a benchmark designed to test simple reasoning in AI models, intended to evaluate the performance of AI on problems that are solvable by at least 10% of high schoolers with a pen, unlimited paper, and an hour of time. The current version includes a preliminary subset of 225 problems.
提供机构：
Salesforce
