Patch Correctness Assessment: A Survey DOI Open Access

Zelin Fei,

Jidong Ge, Chuanyi Li

et al.

ACM Transactions on Software Engineering and Methodology, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 8, 2024

Most automated program repair methods rely on test cases to determine the correctness of generated patches. However, due incompleteness available suites, some patches that pass all may still be incorrect. This issue is known as patch overfitting problem. Overfitting problem a longstanding in repair. Due patches, obtained by tools require further validation their correctness. Researchers have proposed many automatically assess but no systematic review provides detailed introduction this problem, existing solutions, and challenges. To address deficiency, we systematically approaches assessment. We first offer few examples acquire more understanding then propose comprehensive categorization publicly techniques datasets, examine commonly used evaluation metrics, perform an in-depth analysis effectiveness models addressing challenge overfitting. Based our analysis, provided difficulties encountered current methodologies, alongside possible avenues for future research exploration.

Language: Английский

A method to identify overfitting program repair patches based on expression tree DOI

Yukun Dong,

Xiaotong Cheng,

Yufei Yang

et al.

Science of Computer Programming, Journal Year: 2024, Volume and Issue: 235, P. 103105 - 103105

Published: March 5, 2024

Language: Английский

Citations

0

Verification of Programs with Common Fragments DOI
Ivan Postolski, Vı́ctor Braberman, Diego Garbervetsky

et al.

Published: July 10, 2024

Language: Английский

Citations

0

Enhancing the Efficiency of Automated Program Repair via Greybox Analysis DOI
YoungJae Kim,

Yechan Park,

Seungheon Han

et al.

Published: Oct. 18, 2024

Language: Английский

Citations

0

Patch Correctness Assessment: A Survey DOI Open Access

Zelin Fei,

Jidong Ge, Chuanyi Li

et al.

ACM Transactions on Software Engineering and Methodology, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 8, 2024

Most automated program repair methods rely on test cases to determine the correctness of generated patches. However, due incompleteness available suites, some patches that pass all may still be incorrect. This issue is known as patch overfitting problem. Overfitting problem a longstanding in repair. Due patches, obtained by tools require further validation their correctness. Researchers have proposed many automatically assess but no systematic review provides detailed introduction this problem, existing solutions, and challenges. To address deficiency, we systematically approaches assessment. We first offer few examples acquire more understanding then propose comprehensive categorization publicly techniques datasets, examine commonly used evaluation metrics, perform an in-depth analysis effectiveness models addressing challenge overfitting. Based our analysis, provided difficulties encountered current methodologies, alongside possible avenues for future research exploration.

Language: Английский

Citations

0