Using Pre-Trained Models to Boost Code Review Automation

Review Computer Science, Artificial Intelligence

WhoReview: A multi-objective search-based approach for code reviewers recommendation in modern code review

Moataz Chouchen et al.

Summary: In contemporary software development, code review is a crucial practice to ensure software quality. Peer code review is widely adopted, but the manual selection of peer reviewers can be time-consuming and inefficient, leading to increased development costs and time to market. A multi-objective search-based approach, called WhoReview, has been introduced to optimize the selection of peer reviewers for code changes, showing improved results compared to existing approaches.

APPLIED SOFT COMPUTING (2021)

添加到收藏夹

Proceedings Paper Computer Science, Software Engineering

On Multi-Modal Learning of Editing Source Code

Saikat Chakraborty et al.

Summary: Neural Machine Translator (NMT) shows promise in automatically editing source code, but the correct patch depends on developers' requirements and context. Leveraging multiple modalities of information, incorporating developers' hints can improve the accuracy of generating correctly patched code.

2021 36TH IEEE/ACM INTERNATIONAL CONFERENCE ON AUTOMATED SOFTWARE ENGINEERING ASE 2021 (2021)

添加到收藏夹

Proceedings Paper Computer Science, Software Engineering

Towards Automating Code Review at Scale

Vincent J. Hellendoorn et al.

Summary: This study examines the vision and challenges of automating code review at a realistic scale, showing that even predicting the location of comments is challenging due to the tremendous variation and apparent randomness in code reviews. The findings provide a research agenda for realistically and impactfully automating code review.

PROCEEDINGS OF THE 29TH ACM JOINT MEETING ON EUROPEAN SOFTWARE ENGINEERING CONFERENCE AND SYMPOSIUM ON THE FOUNDATIONS OF SOFTWARE ENGINEERING (ESEC/FSE '21) (2021)

添加到收藏夹

Proceedings Paper Computer Science, Software Engineering

Studying the Usage of Text-To-Text Transfer Transformer to Support Code-Related Tasks

Antonio Mastropaolo et al.

Summary: Deep learning techniques, particularly the T5 model, have gained attention in the software engineering community for their ability to support code-related tasks. By pre-training on a large dataset and fine-tuning on specialized data, T5 has shown improved performance in NLP tasks. This paper empirically investigates the performance of T5 in code-related tasks, demonstrating enhancements over previous DL-based solutions.

2021 IEEE/ACM 43RD INTERNATIONAL CONFERENCE ON SOFTWARE ENGINEERING (ICSE 2021) (2021)

添加到收藏夹

Article Computer Science, Software Engineering