Binary and Source Code Function Pairs

该数据集包含27万个二进制函数及其对应源代码函数的对，用于评估二进制逆向工程任务。具体来说，数据集被划分为26万个样本用于训练，1万个样本用于测试，而训练数据的5%则用于验证。规模上，数据集包含27万对函数。任务方面，主要包括从反编译代码中恢复函数名称以及从反编译代码中总结函数语义。

This dataset contains 270,000 pairs of binary functions and their corresponding source code functions, which is used to evaluate binary reverse engineering tasks. Specifically, the dataset is split into 260,000 training samples, 10,000 testing samples, and 5% of the training data is reserved for validation. In terms of scale, the dataset consists of 270,000 function pairs. The main tasks include recovering function names from decompiled code and summarizing function semantics from decompiled code.