一句话：将 LLM 评审员嵌入 GitHub PR 工作流，跨两届课程验证其能规模化支撑代码评审学习并缓解过度依赖。
代码评审是软件工程教育核心，但在毕业设计课程中难以规模化：截止期紧、同伴反馈不均、学生经验不足。本文报告一个直接集成到 GitHub pull request 的 human-in-the-loop LLM 评审员设计，并在 2023 与 2024 两届队列（超过 100 名学生）中部署。研究采用混合方法，结合 GitHub 数据、反思报告与定向问卷，以参与度与响应性作为自我调节学习的行为指标。结果显示：2024 队列迭代活动更多（1176 对 581 个 PR），2023 年的 227 次 AI 评审失败在工具与教学改进后降为零；采用率虽不同（93% 对 50%），响应性却稳定（32% 与 33% 的成功 AI 评审 PR 随后有新提交）。定性上，学生用结构化评论聚焦评审并讨论代码质量，指导性设计减少了过度依赖。贡献为工作流内 AI 评审设计、跨队列比较、混合方法分析与教学建议。
Code review is central to software engineering education but hard to scale in capstones due to tight deadlines, uneven peer feedback, and limited prior experience. We investigate an LLM-as-reviewer integrated directly into GitHub pull requests (human-in-the-loop) across two cohorts (>100 students, 2023–2024). Using a mixed-methods design—GitHub data, reflective reports, and a targeted survey—we examine engagement and responsiveness as behavioural indicators of self-regulated learning processes. Quantitatively, the 2024 cohort produced more iterative activity (1176 vs. 581 PRs), while technical issues observed in 2023 (227 failed AI attempts) dropped to zero after tool and instructional refinements. Despite different adoption levels (teams using the tool: 93% vs. 50%), responsiveness was stable: 32% (2023) and 33% (2024) of successfully AI-reviewed PRs were followed by subsequent commits on the same PR. Qualitatively, students used the LLM’s structured comments to focus reviews and discuss code quality, while guidance reduced over-reliance. We contribute: (i) an in-workflow design for an AI reviewer that scaffolds learning while mitigating cognitive offloading; (ii) a repeated cross-sectional comparison across two cohorts in authentic settings; (iii) a mixed-methods analysis combining objective GitHub metrics with student self-reports; and (iv) evidence-based pedagogical recommendations for responsible, student-led AI-assisted review.
@inproceedings{paperbot12018,
title = {AI-Assisted Code Review as a Scaffold for Code Quality and Self-Regulated Learning: An Experience Report},
author = {Eduardo Oliveira and Michael Fu and Patanamon Thongtanunam and Sonsoles López-Pernas and Mohammed Saqr},
booktitle = {ICSE},
year = {2026},
doi = {10.1145/3786580.3786956}
}