这篇论文提出了 ARCTIC，专门解决 AI 写代码太多、人工审不过来的问题。它用意图预测和漂移检测揪出真正需要人看的改动，比传统工具更准更省 token。
ARCTIC 是一个面向 AI 生成代码的代码批判系统，核心能力包括意图预测、漂移检测和代码聚焦。系统基于 1.8 万条真实代码评审归纳出六主题分类法。离线评测中，意图预测 F1 达 0.86，漂移检测与人工标注的 QWK 为 0.907，代码聚焦在质量估计上比基线 AI 评审员好 2.4 倍且 token 消耗减少 5 倍。试点中漂移评分让代码错位额外降低 5.76 个点（p=0.026），意图预测获 90.2% 认可。
From Code Review to Code Critique: Intent, Drift, and Spotlight for AI-Generated Diffs at Scale
AI coding agents are generating code at volumes that exceed the capacity of traditional peer review. At the same time, existing AI code review tools over-index on low-value suggestions such as style and best practices while under-indexing on the concerns human reviewers prioritize most: correctness, security, and performance. We present ARCTIC, an AI-powered Code Critique system that reframes code review around three capabilities: intent prediction, which infers why a change was made from conversation logs and metadata; drift detection, which measures divergence between the developer's intent and the agent's output via backtranslation; and code spotlight, which ranks the regions of a diff most warranting human scrutiny. We ground these capabilities in a six-theme taxonomy derived from 18,000 code reviews. Offline evaluation shows that intent prediction achieves 0.86 F1, drift detection reaches near-perfect ordinal agreement with human annotators (QWK = 0.907), and spotlight outperforms the baseline AI reviewer by 2.4x on quality estimation at 5x fewer tokens. In the experimental rollout, the drift scores reduces code misalignment by an additional 5.76 points (p = 0.026), intent prediction receives 90.2% approval, and zero defects have been attributed to self-reviewed diffs since launch.