Holger Cleve, Andreas Zeller
A program fails. Under which circumstances does this failure occur? One single algorithm, the delta debugging algorithm, suffices to determine these failure-inducing circumstances. Delta debugging tests a program systematically and automatically to isolate failure-inducing circumstances such as the program input, changes to the program code, or executed statements.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
继续探索
In this paper, we present a novel resilience improvement system for Java applications. The unique feature of this system is to combine automated monitoring, automated perturbation injection, and automated resilience improvement. The latter is achieved thanks to the failure-oblivious computing, a concept introduced in 2004 by Rinard and colleagues. We design and implement the system as agents for the Java virtual machine. We evaluate the system on two real-world applications: a file transfer client and an email server. Our results show that it is possible to automatically improve the resilience…
Modern distributed applications are moving toward a microservice architecture, in which each service is developed and managed independently, and new features and updates are delivered continuously. A guiding principle of microservice architecture is that it must be built to anticipate and mitigate a variety of hardware and software failures. In order to test the fault handling capabilities of microservces, this paper presents IntelliFT, a feedback-based, automated failure testing technique for microservice based applications, which aims to expose the defects in the fault-handling logic quickly…
TrajAudit: Automated Failure Diagnosis for Agentic Coding Systems
针对仓库级编码轨迹中噪声多、上下文长的问题，提出TrajAudit框架，通过模式匹配过滤和测试报告先验知识辅助，实现高效故障诊断，在RootSE基准上定位准确率提升24.4个百分点，令牌消耗降低18%。
The software engineering field has a long history of classifying software failure causes. Understanding them is paramount for fault injection, focusing testing efforts or reliability prediction. Since software fails in manifold complex ways, a broad range of software failure cause models is meanwhile published in dependability literature. We present the results of a meta-study that classifies publications containing a software failure cause model in topic clusters. Our results structure the research field and can help to identify gaps. We applied the systematic mapping methodology for performi…
Easy over Hard: A Simple Baseline for Test Failures Causes Prediction
本文提出NCChecker方法，通过日志抽象、查找表构建和失败原因预测三个阶段自动识别测试失败原因，在超过10K条日志的工业数据集上验证了其高效性和有效性。
CI-Repair-Bench: A Repository-Aware Benchmark for Automated Patch Validation via CI Workflows
CI-Repair-Bench通过真实GitHub Actions执行构建，提供仓库级程序修复基准，包含567个CI失败实例，细粒度评估12种CI错误类型，揭示自动化修复在局部和工具强制失败上效果最佳，但环境依赖等问题仍具挑战。
The momentum gained by microservices and cloud-native software architecture pushed nowadays enterprise IT towards multi-service applications. The proliferation of services and service interactions within applications, often consisting of hundreds of interacting services, makes it harder to detect failures and to identify their possible root causes, which is on the other hand crucial to promptly recover and fix applications. Various techniques have been proposed to promptly detect failures based on their symptoms, viz., observing anomalous behaviour in one or more application services, as well …
Automated debugging techniques, such as Fault Localisation (FL) or Automated Program Repair (APR), are typically designed under the Single Fault Assumption (SFA). However, in practice, an unknown number of faults can independently cause multiple test case failures, making it difficult to allocate resources for debugging and to use automated debugging techniques. Clustering algorithms have been applied to group the test failures according to their root causes, but their accuracy can often be lacking due to the inherent limits in the distance metrics for test cases. We introduce a new test dista…
FlakeRanker: Automated Identification and Prioritization of Flaky Job Failure Categories
本文介绍 FLAKERANKER 工件，提供正则标注脚本、46 类不稳定作业失败的完整成本排名与演变分析，以及 RFM 聚类优先级排序数据集、脚本和 CLI 工具。
Automated Detection and Mitigation of Dependability Failures in Healthcare Scenarios through Digital Twins
本文提出M-GENGAR方法，通过数字孪生闭环范式，实现医疗CPSs中依赖性故障的自动化检测与缓解，利用随机混合自动机建模、数据驱动学习和统计模型检验，有效稳定患者生命体征。
PracRepair: LLM-Empowered Automated Program Repair Inspired by Human-Like Debugging Practices
提出PracRepair框架，通过构建按需静态-动态上下文、进行问题驱动的故障诊断并迭代细化补丁，利用动态信息提升LLM在程序修复中的效果，在Defects4J和真实世界漏洞上取得最优性能。
Stalled, Biased, and Confused: Uncovering Reasoning Failures in LLMs for Cloud-Based Root Cause Analysis
本文通过实证评估揭示LLM在基于云的根本原因分析中的推理失败，评估六种LLM在两种工作流程下的表现，识别16种常见推理失败类型，为未来推理驱动的系统诊断提供指导。
Processing and reviewing nightly test execution failure logs for large industrial systems is a tedious activity. Furthermore, multiple failures might share one root/common cause during test execution sessions, and the review might therefore require redundant efforts. This paper presents the LogGrouper approach for automated grouping of failure logs to aid root/common cause analysis and for enabling the processing of each log group as a batch. LogGrouper uses state-of-art natural language processing and clustering approaches to achieve meaningful log grouping. The approach is evaluated in an in…
Automated Program Repair (APR) techniques typically exploit spectrum-based fault localization (SBFL) to identify the program locations that should be patched, making the effectiveness of APR techniques dependent on the effectiveness of fault localization. Indeed, results show that SBFL often does not localize faults accurately, hindering the effectiveness of APR. In this paper, we propose EXCEPT, a technique that addresses the localization problem by focusing on the semantics of failures rather than on the correlation between the executed statements and the failed tests, as SBFL does. We focus…
L4: Diagnosing Large-scale LLM Training Failures via Automated Log Analysis
本文提出L4框架，通过自动化分析大规模LLM训练日志，识别故障指示信息，实现高效准确的故障诊断，减少人工成本。
ProbeLLM: Automating Principled Diagnosis of LLM Failures
提出ProbeLLM框架，通过分层蒙特卡洛树搜索在全局探索与局部细化间分配预算，结合工具增强生成与验证，将故障发现从孤立案例提升为结构化故障模式，揭示更广泛、清晰、细粒度的故障景观。
The purpose of this study is the representation of Common Cause Failures (CCF) in large digital systems. The system under study is representative of a control system of a nuclear plant. The model for CCF is the generalized Atwood model. It can represent independent failures, CCF non-lethal for some system elements and CCF lethal to all. The Atwood model was modified to "direct" non-lethal DCC on certain parts of the system and take into account the different possible origins of DCC. Maintenance and repairs are taken into account in the model that is thus dynamic. The main evaluation results ar…
An Empirical Study on Failures in Automated Issue Solving
本文对自动化问题解决中的失败进行实证研究，通过分析150个失败实例构建失败模式分类法，并提出协作式Expert-Executor框架，解决22.2%此前无法解决的问题。
Understanding Automated Web GUI Testing: An Empirical Study Across Exploration Strategies and State Abstractions
通过实证研究分析自动化Web GUI测试中探索策略（基于模型、强化学习、大语言模型）与状态抽象如何共同影响测试有效性，发现不同策略在代码覆盖、状态覆盖和故障发现上互补，状态抽象是关键因素。
LLM-Based Automated Diagnosis Of Integration Test Failures At Google
本文提出Auto-Diagnose工具，利用LLM高效诊断集成测试故障，通过分析日志生成摘要并集成至代码审查系统，实验证明其高准确率和用户认可。
Studying and Understanding the Effectiveness and Failures of Conversational LLM-Based Repair
本文以ChatRepair为例，系统研究对话式大语言模型程序修复的有效性与失败原因，通过比较修复策略、评估迭代改进组件和分析失败案例，为未来研究提供关键启示。
BARO: Robust Root Cause Analysis for Microservices via Multivariate Bayesian Online Change Point Detection
针对现有微服务根因分析依赖精准异常检测、结果易受偏差影响的问题，提出融合异常检测与根因分析的端到端方法BARO，通过多变量贝叶斯在线变点检测与非参数假设检验提升鲁棒性，在三个基准系统上性能超越现有最优方法。
Automatically Debugging AutoML Pipelines using Maro: ML Automated Remediation Oracle (Extended Version)
本文提出结合 AutoML 与 SMT 的 Maro 工具，用于自动诊断、解释并修复 AutoML 流水线失败，实验表明其通常一次补救即可修复错误且不影响准确率和收敛时间。
ATTAIN: Automated Exploit Failure Analysis through Trace-Driven Diff Analysis
提出ATTAIN框架，通过轨迹驱动差异分析自动评估库版本中的漏洞存在性，结合LLM搜索相关变更，F1达93.24%，优于现有方法。
Breaking the Cycle of Recurring Failures: Applying Generative AI to Root Cause Analysis in Legacy Banking Systems
本文提出将基于知识的GenAI智能体与“五个为什么”技术结合，用于传统银行系统的事件后根因分析，发现约70%被误判为管理或供应商故障的事件实为内部代码问题，并通过扫描5000多个项目验证了方法的有效性。
Reversible debuggers have been developed at least since 1970. Such a feature is useful when the cause of a bug is close in time to the bug manifestation. When the cause is far back in time, one resorts to setting appropriate breakpoints in the debugger and beginning a new debugging session. For these cases when the cause of a bug is far in time from its manifestation, bug diagnosis requires a series of debugging sessions with which to narrow down the cause of the bug. For such "difficult" bugs, this work presents an automated tool to search through the process lifetime and locate the cause. As…
Automated UI testing is an important component of the continuous integration process of software development. A modern web-based UI is an amalgam of reports from dozens of microservices written by multiple teams. Queries on a page that opens up another will fail if any of that page's microservices fails. As a result, the overall cost for automated UI testing is high since the UI elements cannot be tested in isolation. For example, the entire automated UI testing suite at LexisNexis takes around 30 hours (3-5 hours on the cloud) to execute, which slows down the continuous integration process. T…
Demystifying the Lifecycle of Failures in Platform-Orchestrated Agentic Workflows
本文通过分析平台编排代理工作流中的故障生命周期，提出AgentFail数据集，揭示关键故障机制并提供修复指南。
Containing the Reproducibility Gap: Automated Repository-Level Containerization for Scholarly Jupyter Notebooks
本文提出自动化可重复性工程流水线，通过依赖推断、容器生成和隔离执行，评估学术笔记本的执行环境，解决依赖问题并提升执行稳定性，但发现仍存在低输出保真度的可重复性差距。
This paper presents an approach for modeling software common cause failures (CCFs) within digital instrumentation and control (I&C) systems. CCFs consist of a concurrent failure between two or more components due to a shared failure cause and coupling mechanism. This work emphasizes the importance of identifying software-centric attributes related to the coupling mechanisms necessary for simultaneous failures of redundant software components. The groups of components that share coupling mechanisms are called common cause component groups (CCCGs). Most CCF models rely on operational data as the…
Runtime nondeterminism is a fact of life in modern database applications. Previous research has shown that nondeterminism can cause applications to intermittently crash, become unresponsive, or experience data corruption. We propose Adaptive Interventional Debugging (AID) for debugging such intermittent failures. AID combines existing statistical debugging, causal analysis, fault injection, and group testing techniques in a novel way to (1) pinpoint the root cause of an application's intermittent failure and (2) generate an explanation of how the root cause triggers the failure. AID works by f…
Scaling Mobile Chaos Testing with AI-Driven Test Execution
研究大规模分布式系统中移动应用易受后端服务故障影响的问题，核心方法是将DragonCrawl与uHavoc集成实现人工智能驱动的测试执行，主要贡献是识别风险、缩短调试时间，证明在生产规模实现移动弹性验证可行。
Semantic Evolution over Populations for LLM-Guided Automated Program Repair
本文提出EvolRepair框架，通过语义进化算法提升LLM引导的自动程序修复效果，解决修复假设多样性、语义相关修复家族识别等挑战。
Deep Reinforcement Learning for Automated Web GUI Testing
本文提出WebRLED，一种基于深度强化学习的Web应用自动化GUI测试方法，通过网格动作值学习、动作判别器和好奇心驱动奖励模型提升探索效率，在代码覆盖率和故障检测率上优于现有SOTA技术。
Once a failure is observed, the primary concern of the developer is to identify what caused it in order to repair the code that induced the incorrect behavior. Until a permanent repair is afforded, code repair patches are invaluable. The aim of this work is to devise an automated patch generation technique that proceeds as follows: Step1) It identifies a set of failure-causing control dependence chains that are minimal in terms of number and length. Step2) It identifies a set of predicates within the chains along with associated execution instances, such that negating the predicates at the giv…
Progressing from Anomaly Detection to Automated Log Labeling and Pioneering Root Cause Analysis
本文提出日志异常分类法并探索自动化数据标注，研究多种异常检测技术与异常类型的对齐，并展望异常检测后的根因分析，以提升IT系统韧性、运营效率和用户满意度。
RAG-TESTER: Automated End-to-End Testing of Retrieval-Augmented Large Language Models
RagTester是针对RAG系统的自动化端到端测试方法，通过生成测试用例、利用LLM评估结果，在24种配置中20种优于基线，可有效检测RAG的交互故障并支持部署评估。
Detecting performance issues and identifying their root causes in the runtime is a challenging task. Typically, developers use methods such as logging and tracing to identify bottlenecks. These solutions are, however, not ideal as they are time-consuming and require manual effort. In this paper, we propose a method to automate the task of detecting latency outliers using system-level traces and then comparing them to identify the root cause(s). Our method makes use of dependency graphs to show internal interactions between threads and system resources. With these graphs, one can pinpoint where…
ActiveSaddler: Automated Curriculum Learning for Agent Harness Optimization
ActiveSaddler提出将框架优化中的训练课程建模为非平稳老虎机，动态构建失败模式目标并自适应平衡探索与利用，在GAIA2和Terminal-Bench 2.0上显著提升测试Pass@1。
Using STPA in Compliance with ISO 26262 for Developing a Safe Architecture for Fully Automated Vehicles
本文探讨了如何利用STPA方法扩展ISO 26262的安全范围，支持危害分析与风险评估过程，并通过实际应用识别出系统层面的事故、危害和不安全控制行为。
Automated Multi-Source Debugging and Natural Language Error Explanation for Dashboard Applications
本文提出了一种自动化多源调试系统，利用大型语言模型生成自然语言错误解释，以提高仪表盘应用的调试效率和用户体验。
Industrial robots are important machines applied in numerous modern industries that execute repetitive tasks with high accuracy, replacing or supporting dangerous jobs. In this kind of system, with increased complexity in which cost is related to the time the system keeps working, the system must operate with a minimum number of failures. In other words, a quality aspect important in industry is reliability. We hypothesize that Automated Acceptance Testing improves reliability for industrial robot program. We present the research question, the motivation for this study, our hypothesis and futu…
Automated Root Cause Analysis System for Complex Data Products
本文提出ARCAS，一个基于领域特定语言（DSL）的自动化根因分析诊断平台，通过并行执行自动化故障排除指南（Auto-TSGs）并利用LLM优先级排序与采取行动，实现近实时问题检测与缓解，显著缩短缓解时间并节省工程周期。
Graph-Free Root Cause Analysis
PRISM是一种无需依赖图的高效根因分析框架，在真实世界数据集上实现了68%的Top-1准确率，比基线方法提升了258%。
Who Guards the Benchmarks? Automated Auditing of LLM Agent Benchmarks
BenchGuard通过结构化LLM协议交叉验证所有基准制品，发现ScienceAgentBench中的12个问题，并在BIXBench Verified-50子集中准确识别83.3%的问题，证明自动化基准审计的有效性。
Root-Cause Attribution Is a Search Problem: Continual Search for Long-Horizon Agent Failures
针对长时程AI智能体失败诊断中信息稀疏、分布广导致根因归因困难的问题，提出持续搜索迭代框架，在多个基准上显著提升归因性能，并证明有效搜索优于模型规模。
Failure region, where failure-causing inputs reside, has provided many insights to enhance testing effectiveness of many testing methods. Failure region may also provide some important information to support other processes such as software debugging. When a testing method detects a software failure, indicating that a failure-causing input is identified, the next important question is about how to identify the failure region based on this failure-causing input, i.e., Identification of Failure Regions (IFR). In this paper, we introduce a new IFR strategy, namely Search for Boundary (SB), to ide…
A Topology-Aware, Memory-Centric Architecture that Separates Root-Cause Derivation from Root-Cause Explanation
提出OPS CORTEX多智能体原型，通过四层操作记忆分离根本原因推导（基于依赖图与阈值时序）与解释（LLM），解决微服务故障诊断中上下文丢失问题。
AgentTrace: Causal Graph Tracing for Root Cause Analysis in Deployed Multi-Agent Systems
AgentTrace通过构建因果图追踪多智能体系统部署中的根本原因，实现高精度和低延迟的故障诊断，优于启发式和LLM基线方法。
A fail-operational system for highly automated driving must complete the driving task even in the presence of a failure. This requires redundant architectures and a mechanism to reconfigure the system in case of a failure. Therefore, an arbitration logic is used. For functional safety, the switch-over to a fall-back level must be conducted in the presence of any electric and electronic failure. To provide evidence for a safety argumentation in compliance with ISO 26262, verification of the arbitration logic is necessary. The verification process provides confirmation of the correct failure rea…
Unintentional Security Flaws in Code: Automated Defense via Root Cause Analysis
本文针对初级开发者难以定位漏洞根因的问题，提出结合 T5 与图卷积网络的自动化根因分析工具 T5-RCGCN，实验表明其可将代码安全性提升 28.9%，并促进开发者长期安全能力成长。
扫码添加微信好友，提出您的宝贵建议 👇
💡 备注请填写：网站反馈
发表机构由 AI 从论文文本抽取，如发现错误，欢迎反馈修正建议。