failure mode diagnosis

This work addresses a critical limitation in existing root cause analysis methods for anomalies: their failure to distinguish between two fundamentally distinct sources—measurement errors and mechanism shifts—often leading to misdiagnosis. To resolve this, the paper proposes the first causal framework that explicitly models both anomaly types by treating them as implicit interventions on latent “true” variables and observed “measured” variables. A structural causal model (SCM) with latent variables is constructed, and maximum likelihood estimation is employed to simultaneously classify anomaly types and localize root causes. Theoretically, the approach is shown to be identifiable without requiring prior knowledge of the causal graph structure. Empirical evaluations demonstrate state-of-the-art performance in root cause localization, accurate anomaly-type classification, and robustness even when the underlying causal graph is unknown.

This work addresses the limitations of traditional root cause analysis methods in complex systems lacking dependency graphs, where such approaches often erroneously identify components with high anomaly scores as root causes—particularly when fault propagation amplifies anomalies downstream. To overcome this challenge, the authors propose PRISM, a novel framework that, for the first time, offers theoretically grounded root cause localization without requiring a dependency graph. PRISM leverages a component-based system model and the intrinsic properties of anomaly propagation to construct a lightweight inference algorithm capable of identifying true root causes in the absence of topological information. Evaluated on 735 real-world faults across nine datasets, PRISM achieves a Top-1 accuracy of 68%, representing a 258% improvement over the best baseline method, while requiring only 8 milliseconds per diagnosis.

Industrial root cause diagnosis typically relies on manual hypotheses and extensive fault labels, yet existing data-driven methods suffer from poor interpretability and limited generalization. This work proposes AgentRCA, a novel framework that achieves zero-shot, label-free root cause diagnosis for the first time. By integrating data-driven digital twins with tool-augmented large language models, AgentRCA adopts a hypothesis-driven approach to iteratively gather statistical evidence, evaluate competing hypotheses, and construct transparent reasoning chains that explicitly link observed symptoms to underlying physical faults. Evaluated on real-world multiphase flow facilities and large-scale chemical plants, the framework matches the diagnostic performance of fully supervised baselines while offering high interpretability.

To address the challenges of root cause localization in complex software systems—particularly susceptibility to spurious correlations and incomplete domain expertise—this paper proposes a causal graph modeling method that integrates partial domain knowledge. We introduce a novel four-stage framework: (1) initial causal structure learning via PC/GES variants; (2) reliability enhancement of causal edges using graph neural networks; (3) redundancy elimination through counterfactual reasoning; and (4) lightweight domain knowledge injection, enabling analysts to initiate analysis with only localized expert priors. Evaluated on both synthetic and real-world industrial datasets, our approach achieves a 27.3% improvement in root cause localization accuracy and reduces average causal path length by 41%, outperforming state-of-the-art causal discovery and correlation-based methods. The framework has been deployed in a cloud platform’s performance operations system.

In industrial quality inspection, anomaly detection suffers from poor robustness due to high noise levels and sparse defective samples. To address this, we propose Iterative Refinement of Pseudo-labels (IRP), a self-supervised method that alternately evaluates sample credibility and removes misleading instances under feature-space consistency constraints—effectively purifying the training set dynamically without human annotations and generating high-fidelity self-supervised signals. IRP introduces the novel paradigm of “iterative data refinement,” significantly enhancing model robustness against label noise and cross-domain generalization capability. Evaluated on KSDD2 and MVTec AD benchmarks, IRP consistently outperforms existing unsupervised and self-supervised methods. Notably, under high-noise conditions, it achieves substantial improvements in detection accuracy and reduces false positive rates by over 25%.

This work addresses the limitations of root cause analysis (RCA) in real-world microservice systems, where both classical methods and large language models (LLMs) struggle due to the lack of effective reasoning mechanisms over large-scale, multimodal telemetry data—encompassing metrics, logs, and traces. The authors propose a structured multi-agent RCA framework that operates with or without domain knowledge and introduces a novel backward-reasoning agent to distinguish between “reasoning gaps” and “data ambiguity,” revealing that performance bottlenecks stem from insufficient model reasoning capabilities rather than data scarcity. The framework further integrates automated rule mining to reduce reliance on manual expert knowledge. Experimental results on the OpenRCA benchmark demonstrate that the proposed approach significantly outperforms existing LLM-based and classical RCA methods, underscoring the critical importance of enhanced reasoning and the synergistic integration of domain knowledge.

This work addresses the unreliability of large language model (LLM)-driven agent workflows, which stems from output nondeterminism, complex node dependencies, and tool heterogeneity, and proposes FlowFixer—a novel framework that introduces symbolic reasoning into automated workflow repair. FlowFixer models execution traces symbolically to generate behavioral specifications, enabling precise fault localization and root cause identification, and dynamically synthesizes targeted repair patches. To reduce verification overhead, it incorporates a multidimensional pre-evaluation mechanism. Experimental evaluation on Dify, Coze, and n8n platforms demonstrates that FlowFixer achieves a repair success rate of 71.3%, outperforming existing methods by 11.9%–27.6%, and improves root cause analysis accuracy by 15.3%–38.8%.

This work addresses the limited interpretability and accountability of large language models (LLMs) in root cause analysis, which hinder their applicability in high-stakes operational settings requiring rigorous evidence chains, hypothesis comparison, and uncertainty handling. The authors propose JustDiag, a diagnostic argumentation engine that introduces, for the first time, an explicit modeling of the diagnostic reasoning process into root cause analysis. JustDiag structures and maintains states such as evidence, findings, competing hypotheses, conflicts, and follow-up checks to enable traceable and auditable inference, complemented by a calibration mechanism that explicitly accounts for uncertainty. Integrating LLMs with a structured reasoning framework, the approach employs a two-tier evaluation protocol to assess both outcome and reasoning quality. Experiments on 66 real-world incidents demonstrate that JustDiag significantly outperforms non-argumentative baselines in both outcome and process scores, exhibiting superior uncertainty retention despite a slightly lower completion rate.

Existing approaches to root cause analysis in Kubernetes often rely on scenario-specific heuristics, lacking auditability and reproducibility. This work proposes a graph-guided root cause analysis agent that systematically encodes operational constraints—such as read-only forensics, propagation-aware diagnosis, and independent validation—into a typed evidence graph and a LangGraph state machine. By integrating large language model reasoning with deterministic graph traversal over OpenTelemetry telemetry data, the method enables an auditable and reproducible diagnostic workflow. Evaluated on 23 scenarios from the ITBench benchmark, the approach improves the F1 score for root cause entities from 0.6087 to 0.9130; even without scenario-specific prompts, it maintains a score of 0.6958, demonstrating both effectiveness and partial generalization capability.

Large language model (LLM) agents often produce elusive silent errors in complex tasks that are difficult to localize. This work proposes a closed-loop diagnostic framework that iteratively refines error localization by controllably replaying execution trajectories, diagnosing candidate erroneous steps, applying targeted intervention patches, and leveraging outcome comparisons before and after intervention as attribution evidence. Notably, this approach is the first to directly utilize intervention feedback to refine the attribution process itself, substantially enhancing both accuracy and actionability. Evaluated on four cross-domain multi-hop reasoning benchmarks, the method achieves state-of-the-art error localization performance, particularly excelling in structured tool-use scenarios, and enables effective attribution even in the absence of ground-truth answers.