BenchMark·Hub

hendrycks-MATH-benchmark

B待确认Benchmark
数学竞赛数学MATH 家族开放

发布方:nlile

热度40.0▼ 0.1
下载量 · 30天
14,562
Hugging Face
GitHub Stars
代码仓库
论文被引
6,224
Semantic Scholar
跑分模型 · 30天
Leaderboard results

简介

Hendrycks MATH Dataset Dataset Description The MATH dataset is a collection of mathematics competition problems designed to evaluate mathematical reasoning and problem-solving capabilities in computational systems. Containing 12,500 high school competition-level mathematics problems, this dataset is notable for including detailed step-by-step solutions alongside each problem. Dataset Summary The dataset consists of mathematics problems spanning multiple… See the full description on the dataset page: https://huggingface.co/datasets/nlile/hendrycks-MATH-benchmark.