PCC dataset

PCC数据集是由伊利诺伊大学厄巴纳-香槟分校的研究团队创建，旨在重新评估现有的补丁正确性检查（PCC）技术。该数据集包含1988个补丁，这些补丁由最近的PraPR自动修复工具生成，该工具能够全面探索预定义的补丁搜索空间。数据集的创建是为了解决现有PCC研究中外部有效性的严重威胁，因为用于生成补丁的APR工具只能探索补丁搜索空间的一小部分。通过这个大规模的手工标注数据集，研究团队能够揭示现有PCC技术的各种惊人发现，并为未来PCC技术的发展提供指导和建议。The PCC dataset was created by a research team from the University of Illinois Urbana-Champaign, with the goal of re-evaluating existing patch correctness checking (PCC) techniques. This dataset contains 1988 patches generated by the recent PraPR automated program repair tool, which can comprehensively explore the predefined patch search space. The dataset was developed to address the severe threat to external validity in existing PCC research, as prior APR tools used for patch generation could only explore a small subset of the patch search space. Leveraging this large-scale manually annotated dataset, the research team was able to uncover various surprising findings about existing PCC technologies, and provide actionable guidance and recommendations for the future development of PCC techniques.

搜集汇总

背景与挑战

背景概述

PCC数据集是一个用于评估补丁正确性检查技术的大规模数据集，包含1988个由PraPR自动修复工具生成的补丁，旨在解决现有PCC研究的外部有效性威胁，并通过手工标注揭示技术发现，为未来PCC发展提供建议。