一句话：提出现实评测设定RDS，发现深度学习不如传统IR，并用向量检索与LLM重排实现EasyLink。
Issue-commit链接对软件维护至关重要，但现有评测忽略了仓库中大量相似无关提交的干扰。作者提出Realistic Distribution Setting（RDS），构建含20个开源项目的现实评测数据集。结果显示，最先进深度学习方法性能下降超一半，传统VSM反而更优。受此启发，作者提出EasyLink：以向量数据库进行现代信息检索，并用大语言模型对检索到的提交重排序，以缓解issue与commit的语义鸿沟。评测中EasyLink平均Precision@1达75.03%，较最先进方法提升四倍以上。论文还给出推进该方向研究的实践指南。
Issue-commit linking, which connects issues with commits that fix them, is crucial for software maintenance. Existing approaches have shown promise in automatically recovering these links. Evaluations of these techniques assess their ability to identify genuine links from plausible but false links. However, these evaluations overlook the fact that, in reality, when a repository has more commits, the presence of more plausible yet unrelated commits may interfere with the tool in differentiating the correct fix commits. To address this, we propose the Realistic Distribution Setting (RDS) and use it to construct a more realistic evaluation dataset that includes 20 open-source projects. By evaluating tools on this dataset, we observe that the performance of the state-of-the-art deep learning-based approach drops by more than half, while the traditional Information Retrieval method, VSM, outperforms it. Inspired by these observations, we propose EasyLink, which utilizes a vector database as a modern Information Retrieval technique. To address the long-standing problem of the semantic gap between issues and commits, EasyLink leverages a large language model to rerank the commits retrieved from the database. Under our evaluation, EasyLink achieves an average Precision@1 of 75.03%, improving over the state-of-the-art by over four times. Additionally, this paper provides practical guidelines for advancing research in issue-commit link recovery.
@inproceedings{abs-2507-09199,
title = {Back to the Basics: Rethinking Issue-Commit Linking with LLM-Assisted Retrieval},
author = {Huihui Huang and Ratnadira Widyasari and Ting Zhang and Ivana Clairine Irsan and Jieke Shi and Han Wei Ang and Frank Liauw and Eng Lieh Ouh and Lwin Khin Shar and Hong Jin Kang and David Lo},
booktitle = {ICSE},
year = {2026},
doi = {10.1145/3744916.3764530}
}