对联小数据集

对联的小数据集，用于验证模型上下句生成的泛化能力，本数据集包括训练集（29w）测试集（2w）。其中，每一条数据由一副对联组成，上半是输入text1，输出句子text2，{"text1":"妻贤家富贵","text2":"心静梦清凉"}

A small dataset of Chinese couplets, designed to validate the generalization capability of models for generating the second line of a couplet given the first line. This dataset comprises a training set (290,000 samples) and a test set (20,000 samples). Each entry in the dataset consists of a complete Chinese couplet, with the first line as the input text1 and the second line as the target output text2. An example entry is: {"text1":"妻贤家富贵","text2":"心静梦清凉"}

搜集汇总

背景与挑战

背景概述

对联小数据集是一个中文文本生成数据集，旨在测试模型根据上联生成下联的泛化能力。该数据集采用Apache 2.0许可证，于2022年12月发布，大小为14.64MB。