Miku Watanabe

Miku Watanabe

Auto-generated profile, based on publicly available information

Miku Watanabe

Nara Institute of Science and Technology, Japan

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 March 2026 in ACM Transactions on Software Engineering and Methodology
Verified
Journal Article
Open access
MWMiku Watanabe
Miku Watanabe
HLHao Li
Hao Li
YKYutaro Kashiwa
Yutaro Kashiwa
BRBrittany Reid
Brittany Reid
HIHajimu Iida
Hajimu Iida
AHAhmed E. Hassan
Ahmed E. Hassan
Large language models (LLMs) are increasingly being integrated into software development processes. The ability to generate code and submit pull requests with minimal human intervention, through the use of autonomous AI agents, is poised to become a standard practice. However, little is known about the practical usefulness of these pull requests and the extent to which their contributions are accepted in real-world projects. In this paper, we empirically study 567 GitHub pull requests (PRs) generated using Claude Code, an agentic coding tool, across 157 diverse open-source projects. Our analysis reveals that developers tend to rely on agents for tasks such as refactoring, documentation, and testing. The results indicate that 83.8% of these agent-assisted PRs are eventually accepted and merged by project maintainers, with 54.9% of the merged PRs being integrated without further modification. The remaining 45.1% require additional changes and benefit from human revisions, especially for bug fixes, documentation, and adherence to project-specific standards. These findings suggest that while agent-assisted PRs are largely acceptable, they still benefit from human oversight and refinement.

18 June 2024
Verified
Conference Paper
Open access
MWMiku Watanabe
Miku Watanabe
YKYutaro Kashiwa
Yutaro Kashiwa
BLBin Lin
Bin Lin
THToshiki Hirao
Toshiki Hirao
KYKen'IChi Yamaguchi
Ken'IChi Yamaguchi
HIHajimu Iida
Hajimu Iida
Code review is a critical but time-consuming process for ensuring code quality in modern software engineering. To alleviate the effort of reviewing source code, recent studies have investigated the possibility of automating the review process. Moreover, tools based on large language models such as ChatGPT are playing an increasingly important role in this vision. Understanding how these tools are used during code review can provide valuable insights for code review automation. This study investigates for what purposes developers use ChatGPT during code review and how developers react to the information and suggestions provided by ChatGPT. We manually analyze 229 review comments in 205 pull requests from 179 projects. We find that developers often use ChatGPT for outsourcing their work as frequently as asking for references. Moreover, we observe that only 30.7% of responses to the answers provided by ChatGPT are negative. We further analyze the reasons behind the negative reactions. Our results provide valuable insights for improving the effectiveness of LLMs in code reviews.

12 April 2026
Verified
Conference Paper
Open access
AHAhmed E. Hassan
Ahmed E. Hassan
HLHao Li
Hao Li
DLDayi Lin
Dayi Lin
BABram Adams
Bram Adams
TCTse-Hsun (Peter) Chen
Tse-Hsun (Peter) Chen
MWMiku Watanabe
Miku Watanabe