一句话:提出 CoCoNuT,通过集成多个上下文感知的神经翻译模型来修复程序,克服传统 APR 依赖硬编码规则的局限。
自动生成-验证(GV)程序修复(APR)技术通常依赖硬编码规则,只能修复符合特定修复模式的缺陷,规则发现需大量人工,且难以适配不同编程语言。CoCoNuT 提出一种基于集成(ensemble)的方法,结合多个上下文感知的神经翻译模型进行程序修复。该方法利用神经翻译从上下文中学习修复模式,以数据驱动方式替代手工规则,旨在提高修复的通用性和跨语言能力。摘要未给出具体实验评估结果。
Automated generate-and-validate (GV) program repair techniques (APR) typically rely on hard-coded rules, thus only fixing bugs following specific fix patterns. These rules require a significant amount of manual effort to discover and it is hard to adapt these rules to different programming languages.
DOI 原文 ·
@inproceedings{paperbot6076,
title = {CoCoNuT: combining context-aware neural translation models using ensemble for program repair},
author = {Thibaud Lutellier and Hung Viet Pham and Lawrence Pang and Yitong Li and Moshi Wei and Lin Tan},
booktitle = {ISSTA},
year = {2020},
doi = {10.1145/3395363.3397369}
}