一句话：DepRadar 用多个专门化智能体协同静态分析与领域规则，自动判定深度学习库缺陷是否影响下游程序。
深度学习库（如 Transformers、Megatron）被广泛复用，但其缺陷（从静默计算错误到细微性能回退）是否会影响下游用户程序往往难以判断：这既需要理解缺陷语义，又需检查客户端代码是否满足涉及配置开关、运行环境和间接 API 调用的复杂触发条件。本文提出 DepRadar，一个面向深度学习库更新的细粒度缺陷与影响分析的智能体协同框架。它分三步协调四个专门化智能体：PR Miner 与 Code Diff Analyzer 从 commit 或 pull request 中抽取结构化缺陷语义；Orchestrator Agent 将这些信号合成为带触发条件的统一缺陷模式；Impact Analyzer 检查下游程序能否触发该缺陷。为提升准确性与可解释性，DepRadar 将静态分析与深度学习领域规则结合，用于缺陷推理与客户端追踪。在两类代表性深度学习库的 157 个 PR 与 70 个 commit 上，其缺陷识别精确率达 90%，结构化字段平均得分 1.6/2；在 122 个客户端程序上以 90% 召回率、80% 精确率识别受影响案例，显著优于基线。
Deep learning (DL) libraries like Transformers and Megatron are now widely adopted in modern AI programs. However, when these libraries introduce defects—ranging from silent computation errors to subtle performance regressions—it is often challenging for downstream users to assess whether their own programs are affected. Such impact analysis requires not only understanding the defect semantics but also checking whether the client code satisfies complex triggering conditions involving configuration flags, runtime environments, and indirect API usage. We present DepRadar, an agent coordination framework for fine-grained defect and impact analysis in DL library updates. DepRadar coordinates four specialized agents across three steps: (1) the PR Miner and Code Diff Analyzer extract structured defect semantics from commits or pull requests, (2) the Orchestrator Agent synthesizes these signals into a unified defect pattern with trigger conditions, and (3) the Impact Analyzer checks downstream programs to determine whether the defect can be triggered. To improve accuracy and explainability, DepRadar integrates static analysis with DL-specific domain rules for defect reasoning and client-side tracing. We evaluate DepRadar on 157 PRs and 70 commits across two representative DL libraries. It achieves 90% precision in defect identification and generates high-quality structured fields (average field score 1.6/2). On 122 client programs, DepRadar identifies affected cases with 90% recall and 80% precision, substantially outperforming other baselines.
@inproceedings{abs-2601-09440,
title = {Depradar: Agentic Coordination for Context-Aware Defect Impact Analysis in Deep Learning Libraries},
author = {Yi Gao and Xing Hu and Tongtong Xu and Jiali Zhao and Xiaohu Yang and Xin Xia},
booktitle = {ICSE},
year = {2026},
doi = {10.1145/3744916.3787763}
}