ARQMath

该数据集是ArqMATH竞赛提供的一组数学栈交换（MSE）的问答对集合。在竞赛第三版的任务1中，重点是从MSE中检索与78个本科级别数学问题相关的答案。对于排名靠前的答案的相关性，由学生评估者进行评估，评估所使用的指标包括平均精度均值（mAP）、前10个答案的精确度（P@10）以及归一化累积增益（nDCG）。这项任务的目标是从MSE中为数学问题检索相关答案。

This dataset is a collection of question-answer pairs from Mathematics Stack Exchange (MSE) provided by the ArqMATH competition. In Task 1 of the 3rd edition of the competition, the focus is on retrieving relevant answers from MSE for 78 undergraduate-level mathematics questions. The relevance of top-ranked answers is evaluated by student evaluators, using metrics including Mean Average Precision (mAP), Precision@10 (P@10), and Normalized Discounted Cumulative Gain (nDCG). The goal of this task is to retrieve relevant answers for mathematical questions from MSE.

搜集汇总

背景与挑战

背景概述

ARQMath数据集是基于Math Stack Exchange平台构建的数学问题解答数据集，主要用于评估大型语言模型在数学领域的性能。该数据集包含Arqmath3竞赛数据，支持生成答案和嵌入向量，旨在研究模型处理复杂数学问题的能力。