GSM8K-Prolog
收藏arXiv2025-09-30 收录
下载链接：
https://github.com/yuce/pyswip
下载链接
链接失效反馈官方服务：
资源简介：
该数据集名为GSM8K-Prolog，旨在用于训练和评估大型语言模型，使其能够通过生成Prolog程序来解决算术问题。该数据集还包括一个由训练集中的100个样本组成的验证集，并已通过使用排列样本进行了增强，以提高模型性能。数据集的总规模为8792个样本，其中训练样本有7473个，测试样本有1319个，其任务是解决算术问题。
This dataset, named GSM8K-Prolog, is designed for training and evaluating large language models (LLMs) to solve arithmetic problems via generating Prolog programs. It additionally includes a validation set comprising 100 samples sourced from the training set, and has been augmented with permuted samples to improve model performance. With a total scale of 8792 samples, the dataset consists of 7473 training samples and 1319 test samples, all focused on solving arithmetic problems.
提供机构：
GSM8K
