Reproduce “Benchmarking and Enhancing Design-Aware Code Review in the Linux Kernel” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Zhenhao Zhou, Kai Yu, Ying Wang, Zhiqiang Yuan, Junhao Zeng, Yijian Wu, Chong Wang, Xin Peng
DesignLore identifies documented design trade-offs in code review by organizing past discussions into decision cases, retrieving relevant cases for a patch, and checking their conditions against current code. On a benchmark of 72 Linux memory-management patch series and 79 trade-offs, it achieves an F1 score of 0.259, versus 0.213 for Sashiko; this measures documented trade-off detection, not general code correctness.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.