Flow-of-action: SOP-enhanced LLM-based multi-agent system for root cause analysis
Open accessAbstract
In the realm of microservices architecture, the occurrence of frequent incidents necessitates the employment of Root Cause Analysis (RCA) for swift issue resolution. It is common that a serious incident can take several domain experts hours to identify the root cause. Consequently, a contemporary trend involves harnessing Large Language Models (LLMs) as automated agents for RCA. Though the recent ReAct framework aligns well with the Site Reliability Engineers (SREs) for its thought-action-observation paradigm, its hallucinations often lead to irrelevant actions and directly affect subsequent results. Additionally, the complex and variable clues of the incident can overwhelm the model one step further. To confront these challenges, we propose Flow-of-Action, a pioneering Standard Operation Procedure (SOP) enhanced LLM-based multi-agent system. By explicitly summarizing the diagnosis steps of SREs, SOP imposes constraints on LLMs at crucial junctures, guiding the RCA process towards the correct trajectory. To facilitate the rational and effective utilization of SOPs, we design an SOP-centric framework called SOP flow. SOP flow contains a series of tools, including one for finding relevant SOPs for incidents, another for automatically generating SOPs for incidents without relevant ones, and a tool for converting SOPs into code. This significantly alleviates the hallucination issues of ReAct in RCA tasks. We also design multiple auxiliary agents to assist the main agent by removing useless noise, narrowing the search space, and informing the main agent whether the RCA procedure can stop. Compared to the ReAct method's 35.50% accuracy, our Flow-of-Action method achieves 64.01%, meeting the accuracy requirements for RCA in real-world systems.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Budgets prevent open-ended tool churn during incidents and provide a mechanical point at which an agent must stop, escalate, or roll back.
“Budgets are not only a cost-control device.They prevent openended tool churn during incidents and provide a mechanical point at which an agent must stop, escalate, or roll back [43], [40], [79].”
From Large language models for agentic NetOps and AIOps: Architectures, evaluation, and safety (2027) · cites this paper as [43]
Section 3.4 (page unavailable) says the authors imposed a 20-step limit to prevent unbounded loops in root-cause analysis, supporting the claim that a budget prevents open-ended activity. The paper does not say that reaching the limit triggers escalation or rollback, or specify what the agent must do at that point. (LLM, conf=0.87)
Experimental — accuracy not yet measuredAn agent may identify the right cause and still fail operationally if it cannot show evidence, pass checks, prepare rollback, or leave an audit trail.
“An agent may identify the right cause and still fail operationally if it cannot show evidence, pass checks, prepare rollback, or leave an audit trail [85], [43], [74].”
From Large language models for agentic NetOps and AIOps: Architectures, evaluation, and safety (2027) · cites this paper as [43]
The Introduction (page unavailable) says that identifying a root cause without showing the diagnostic process leads to poor explainability and SRE distrust. That supports the point about showing how a cause was identified, but our copy does not establish the claimed requirements to pass checks, prepare rollback, or leave an audit trail. (LLM, conf=0.86)
Experimental — accuracy not yet measuredThe reference architecture exposes observables that can be measured across tasks, traces, and deployment stages.
“It exposes observables that can be measured across tasks, traces, and deployment stages [43], [79], [107].”
From Large language models for agentic NetOps and AIOps: Architectures, evaluation, and safety (2027) · cites this paper as [43]
Section 3.1.1 and Appendix A (page numbers unavailable) describe collecting metrics, logs, and traces from a deployed microservices system, supporting the claim about measurable observables and traces. The paper does not describe those observables as spanning tasks or deployment stages, or present its system as the cited document’s reference architecture. (LLM, conf=0.84)
Experimental — accuracy not yet measuredAutonomy changes which failures dominate.
“Autonomy changes which failures dominate [85], [43], [31].”
From Large language models for agentic NetOps and AIOps: Architectures, evaluation, and safety (2027) · cites this paper as [43]
Section 3.2 (page unavailable) reports that smaller action sets restrict randomness but leave the model unable to handle complex scenarios, while larger sets introduce randomness and loss of control. This supports a shift in failure modes as an RCA agent’s action choices change, but the paper does not establish that autonomy changes which failures dominate generally. (LLM, conf=0.78)
Experimental — accuracy not yet measuredSafe autonomy can only be measured when tools, gates, and action envelopes are part of the benchmark.
“These needs go back to Section III: safe autonomy can only be measured when tools, gates, and action envelopes are part of the benchmark [31], [43], [79].”
From Large language models for agentic NetOps and AIOps: Architectures, evaluation, and safety (2027) · cites this paper as [43]
Sections 2.2 and 2.4 describe tools and an action set that constrains agent choices; Section 3.4 evaluates the effects of removing the action set and SOP flow. That supports including these components in this paper’s RCA evaluation, but the paper does not say that safe autonomy can only be measured by a benchmark containing tools, gates, and action envelopes. (LLM, conf=0.75)
Experimental — accuracy not yet measured
RCA quality depends on tool sequencing and stopping rules, not only on language.
“For AIOps, the point is that RCA quality depends on tool sequencing and stopping rules, not only on language [86], [43], [79].”
From Large language models for agentic NetOps and AIOps: Architectures, evaluation, and safety (2027) · cites this paper as [43]
Recent IT automation benchmarks use real tasks and interfaces, not merely static text.
“Recent IT automation benchmarks use real tasks and interfaces, not merely static text [43], [79].”
From Large language models for agentic NetOps and AIOps: Architectures, evaluation, and safety (2027) · cites this paper as [43]
Benchmarks created only from clear incident descriptions are likely to exaggerate safe autonomy.
“Benchmarks created only from clear incident descriptions are likely to exaggerate safe autonomy [43], [79].”
From Large language models for agentic NetOps and AIOps: Architectures, evaluation, and safety (2027) · cites this paper as [43]
The Dataset section (3.1.1; page unavailable) describes a benchmark of 90 injected incidents using metrics, logs, and traces, while Challenge 1 discusses hallucinations and errors during root-cause analysis. Neither passage assesses benchmarks made only from clear incident descriptions or says they exaggerate safe autonomy; searches for those terms found no such assessment. (LLM, conf=0.70)
Experimental — accuracy not yet measuredWithout T and Π, evaluation reduces to story quality.
“Without T and Π, evaluation reduces to story quality [43], [79], [40], [20].”
From Large language models for agentic NetOps and AIOps: Architectures, evaluation, and safety (2027) · cites this paper as [43]
Section 3.1.2 defines root-cause location and type accuracy and average path length as evaluation metrics. Searches for the quoted claim and for evaluation without traces or a policy found no passage saying that, without T and Π, evaluation reduces to story quality. (LLM, conf=0.72)
Experimental — accuracy not yet measuredITBench contains 94 scenarios across SRE, CISO, and FinOps, and its early results report low success rates for current agents, including 13.8% for SRE tasks and 25.2% for CISO tasks.
“It contains 94 scenarios across SRE, CISO, and FinOps, and its early results report low success rates for current agents, including 13.8% for SRE tasks and 25.2% for CISO tasks [43].”
From Large language models for agentic NetOps and AIOps: Architectures, evaluation, and safety (2027) · cites this paper as [43]
The Dataset section (3.1.1; page unavailable) describes 90 Kubernetes incidents, while the performance table (section 3.3; page unavailable) reports root-cause-analysis accuracy metrics. Searches for ITBench, its 94 scenarios, and the stated SRE and CISO success rates found no passage addressing those claims. (LLM, conf=0.76)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Advances in Neural Information Processing Systems 19
(2007) · The MIT Press eBooks · doi:10.52202/085713 · Cited by 16,037 · see citing works
Unsupervised Anomaly Detection via Variational Auto-Encoder for Seasonal KPIs in Web Applications
Xu, H., Feng, Y., Chen, J., Wang, Z., Qiao, H., Chen, W., Zhao, N., Li, Z., Bu, J., Li, Z., et al. (2018) · International World Wide Web Conference (WWW) · doi:10.1145/3178876.3185996 · Cited by 1,001 · see citing works
Self-citationMicroRCA: Root Cause Localization of Performance Issues in Microservices
Wu, L., Tordsson, J., Elmroth, E., & Kao, O. (2020) · NOMS 2020 - 2020 IEEE/IFIP Network Operations and Management Symposium · doi:10.1109/noms47738.2020.9110353 · Cited by 265 · see citing works
Unsupervised Detection of Microservice Trace Anomalies through Service-Level Deep Bayesian Networks
Liu, P., Xu, H., Ouyang, Q., Jiao, R., Chen, Z., Zhang, S., Yang, J., Mo, L., Zeng, J., Xue, W., et al. (2020) · 2020 IEEE 31st International Symposium on Software Reliability Engineering (ISSRE) · doi:10.1109/issre5003.2020.00014 · Cited by 252 · see citing works
Self-citationAutomatic Root Cause Analysis via Large Language Models for Cloud Incidents
Chen, Y., Xie, H., Ma, M., Kang, Y., Gao, X., Shi, L., Cao, Y., Gao, X., Fan, H., Wen, M., et al. (2024) · Proceedings of the Nineteenth European Conference on Computer Systems · doi:10.1145/3627703.3629553 · Cited by 190 · see citing works
Yu, G., Chen, P., Chen, H., Guan, Z., Huang, Z., Jing, L., Weng, T., Sun, X., & Li, X. (2021) · Proceedings of the Web Conference 2021 · doi:10.1145/3442381.3449905 · Cited by 158 · see citing works
Yu, G., Chen, P., Li, Y., Chen, H., Li, X., & Zheng, Z. (2023) · Proceedings of the 31st ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering · doi:10.1145/3611643.3616249 · Cited by 140 · see citing works
Revisiting VAE for Unsupervised Time Series Anomaly Detection: A Frequency Perspective
Wang, Z., Pei, C., Ma, M., Wang, X., Li, Z., Pei, D., Rajmohan, S., Zhang, D., Lin, Q., Zhang, H., et al. (2024) · International World Wide Web Conference (WWW) · doi:10.1145/3589334.3645710 · Cited by 107 · see citing works
Self-citationActionable and interpretable fault localization for recurring failures in online service systems
Li, Z., Zhao, N., Li, M., Lu, X., Wang, L., Chang, D., Nie, X., Cao, L., Zhang, W., Sui, K., et al. (2022) · ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE) · doi:10.1145/3540250.3549092 · Cited by 75 · see citing works
Self-citationD-Bot: Database Diagnosis System using Large Language Models
Zhou, X., Li, G., Sun, Z., Liu, Z., Chen, W., Wu, J., Liu, J., Feng, R., & Zeng, G. (2024) · Proceedings of the VLDB Endowment · doi:10.14778/3675034.3675043 · Cited by 75 · see citing works
Mining Historical Test Logs to Predict Bugs and Localize Faults in the Test Logs
Amar, A., & Rigby, P. C. (2019) · ACM/IEEE International Conference on Software Engineering: Companion Proceedings (ICSE-Companion) · doi:10.1109/icse.2019.00031 · Cited by 74 · see citing works
MULAN: Multi-modal Causal Structure Learning and Root Cause Analysis for Microservice Systems
Zheng, L., Chen, Z., He, J., & Chen, H. (2024) · International World Wide Web Conference (WWW) · doi:10.1145/3589334.3645442 · Cited by 59 · see citing works
Network-Centric Distributed Tracing with DeepFlow: Troubleshooting Your Microservices in Zero Code
Shen, J., Zhang, H., Xiang, Y., Shi, X., Li, X., Shen, Y., Zhang, Z., Wu, Y., Yin, X., Wang, J., et al. (2023) · ACM SIGCOMM Conference (SIGCOMM) · doi:10.1145/3603269.3604823 · Cited by 54 · see citing works
Self-citationRoot Cause Analysis in Microservice Using Neural Granger Causal Discovery
Lin, C.-M., Chang, C., Wang, W.-Y., Wang, K.-D., & Peng, W.-C. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i1.27772 · Cited by 39 · see citing works
Ding, R., Zhang, C., Wang, L., Xu, Y., Ma, M., Wu, X., Zhang, M., Chen, Q., Gao, X., Gao, X., et al. (2023) · ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE) · doi:10.1145/3611643.3613864 · Cited by 38 · see citing works
Zhang, W., Guo, H., Yang, J., Tian, Z., Zhang, Y., Chaoran, Y., Li, Z., Li, T., Shi, X., Zheng, L., et al. (2024) · Findings of the Association for Computational Linguistics: EMNLP 2024 · doi:10.18653/v1/2024.findings-emnlp.232 · Cited by 35 · see citing works
Self-citationCausIL: Causal Graph for Instance Level Microservice Data
Chakraborty, S., Garg, S., Agarwal, S., Chauhan, A., & Saini, S. K. (2023) · International World Wide Web Conference (WWW) · doi:10.1145/3543507.3583274 · Cited by 24 · see citing works
Yao, Z., Pei, C., Chen, W., Wang, H., Su, L., Jiang, H., Xie, Z., Nie, X., & Pei, D. (2024) · Companion Proceedings of the 32nd ACM International Conference on the Foundations of Software Engineering · doi:10.1145/3663529.3663827 · Cited by 21 · see citing works
Self-citationRosenberg, C. M., & Moonen, L. (2020) · Proceedings of the ... ACM-IEEE International Symposium on Empirical Software Engineering and Measurement/International Symposium on Empirical Software Engineering and Measurement · doi:10.1145/3382494.3410684 · Cited by 20 · see citing works
MicroFI: Non-Intrusive and Prioritized Request-Level Fault Injection for Microservice Applications
Chen, H., Chen, P., Yu, G., Li, X., & He, Z. (2024) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2024.3363902 · Cited by 19 · see citing works
GAMMA: Graph Neural Network-Based Multi-Bottleneck Localization for Microservices Applications
Somashekar, G., Dutt, A., Adak, M., Lorido-Botrán, T., & Gandhi, A. (2024) · International World Wide Web Conference (WWW) · doi:10.1145/3589334.3645665 · Cited by 17 · see citing works
Look Deep into the Microservice System Anomaly through Very Sparse Logs
Jiang, X., Pan, Y., Ma, M., & Wang, P. (2023) · International World Wide Web Conference (WWW) · doi:10.1145/3543507.3583338 · Cited by 12 · see citing works
Flagged first, then by citations
2 of the 24 references couldn’t be matched to a known paper.
Research works printed but not linked: 15
- Sirui Hong, Xiawu Zheng, Jonathan Chen, Yuheng Cheng, Jinlin Wang, Ceyao Zhang, Zili Wang, Steven Ka Shing Yau, Zijuan Lin, Liyang Zhou, et al. 2023. Metagpt: Meta programming for multi-agent collaborative framework. arXiv preprint arXiv:2308.00352 (2023). · arXiv
- Soyeong Jeong, Jinheon Baek, Sukmin Cho, Sung Ju Hwang, and Jong C Park. 2024. Adaptive-rag: Learning to adapt retrieval-augmented large language models through question complexity. arXiv preprint arXiv:2403.14403 (2024). · arXiv
- Chengshu Li, Jacky Liang, Andy Zeng, Xinyun Chen, Karol Hausman, Dorsa Sadigh, Sergey Levine, Li Fei-Fei, Fei Xia, and Brian Ichter. 2023. Chain of code: Reasoning with a language model-augmented code emulator. arXiv preprint arXiv:2312.04474 (2023). · arXiv
- Panagiotis Misiakos, Chris Wendler, and Markus Püschel. 2024. Learning DAGs from data with few root causes. Advances in Neural Information Processing Systems 36 (2024).
- Liangming Pan, Alon Albalak, Xinyi Wang, and William Yang Wang. 2023. Logic- lm: Empowering large language models with symbolic solvers for faithful logical reasoning. arXiv preprint arXiv:2305.12295 (2023). · arXiv
- Yujia Qin, Shihao Liang, Yining Ye, Kunlun Zhu, Lan Yan, Yaxi Lu, Yankai Lin, Xin Cong, Xiangru Tang, Bill Qian, et al. 2023. Toolllm: Facilitating large language models to master 16000+ real-world apis. arXiv preprint arXiv:2307.16789 (2023). · arXiv
- Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, Ilya Sutskever, et al. 2019. Language models are unsupervised multitask learners. OpenAI blog 1, 8 (2019), 9.
- Timo Schick, Jane Dwivedi-Yu, Roberto Dessì, Roberta Raileanu, Maria Lomeli, Eric Hambro, Luke Zettlemoyer, Nicola Cancedda, and Thomas Scialom. 2024. Toolformer: Language models can teach themselves to use tools. Advances in Neural Information Processing Systems 36 (2024).
- Noah Shinn, Federico Cassano, Ashwin Gopinath, Karthik Narasimhan, and Shunyu Yao. 2024. Reflexion: Language agents with verbal reinforcement learning. Advances in Neural Information Processing Systems 36 (2024).
- Shreshth Tuli, Giuliano Casale, and Nicholas R Jennings. 2022. Tranad: Deep transformer networks for anomaly detection in multivariate time series data. arXiv preprint arXiv:2201.07284 (2022). · arXiv
- Lu Wang, Chaoyun Zhang, Ruomeng Ding, Yong Xu, Qihang Chen, Wentao Zou, Qingjun Chen, Meng Zhang, Xuedong Gao, Hao Fan, et al. 2023. Root cause analysis for microservice systems via hierarchical reinforcement learning from human feedback. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 5116-5125.
- Zexin Wang, Jianhui Li, Minghua Ma, Ze Li, Yu Kang, Chaoyun Zhang, Chetan Bansal, Murali Chintalapati, Saravan Rajmohan, Qingwei Lin, et al. 2024. Large Language Models Can Provide Accurate and Interpretable Incident Triage. In 2024 IEEE 35th International Symposium on Software Reliability Engineering (ISSRE). IEEE, 523-534.
- Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, et al. 2022. Chain-of-thought prompting elicits reasoning in large language models. Advances in neural information processing systems 35 (2022), 24824-24837.
- Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik Narasimhan, and Yuan Cao. 2022. React: Synergizing reasoning and acting in language models. arXiv preprint arXiv:2210.03629 (2022). · arXiv
- Zhenhe Yao, Haowei Ye, Changhua Pei, Guang Cheng, Guangpei Wang, Zhiwei Liu, Hongwei Chen, Hang Cui, Zeyan Li, Jianhui Li, et al. 2024. SparseRCA: Unsupervised Root Cause Analysis in Sparse Microservice Testing Traces. In 2024 IEEE 35st International Symposium on Software Reliability Engineering (ISSRE).
Cited by
Sorted by citations
AI-Augmented SOC: A Survey of LLMs and Agents for Security Automation
Srinivas, S., Kirk, B., Zendejas, J., Espino, M., Boskovich, M., Bari, A., Dajani, K., & Alzahrani, N. (2025) · Journal of Cybersecurity and Privacy · doi:10.3390/jcp5040095 · Cited by 39 · see citing works
Triangle: Empowering Incident Triage with Multi-Agent
Yu, Z., Fang, A., Ma, M., Walia, J. S., Zhang, C., Chi, S., Li, Z., Chintalapati, M., Zhang, X., Wang, R., et al. (2025) · IEEE/ACM International Conference on Automated Software Engineering (ASE) · doi:10.1109/ase63991.2025.00062 · Cited by 10 · see citing works
Self-citationVeriTrail-RCA: Evidence-Verified Root Cause Localization for Microservice Backends
Hu, R., Zheng, Y., Fang, R., & Zhou, J. (2026) · International Conference on Machine Learning and Intelligent Systems Engineering (MLISE) · doi:10.1109/mlise70044.2026.11607533 · Cited by 7 · see citing works
VerifyRCA: LLM-Guided but Evidence-Verified Root Cause Analysis for Microservice Backends
Ma, W., Yang, K., Liu, Z., Jiang, R., & Jiang, Y. (2026) · International Conference on Electronic Technology and Information Science (ICETIS) · doi:10.1109/icetis70504.2026.11633469 · Cited by 3 · see citing works
Multi-agent Framework for AI-Supported Collaborative Root Cause Analysis in Quality Assurance
Bocanet, V. I., Muntean, M. H., & Fleșeriu, C. (2025) · IFIP advances in information and communication technology · doi:10.1007/978-3-032-03538-7_15 · Cited by 2 · see citing works
Praxis: Integrating Program Analysis with Observability for Root-Cause Analysis
Cui, S., Krishna, R., Jha, S., & Iyer, R. K. (2026) · International Conference on Dependable Systems & Networks/Proceedings · doi:10.1109/dsn69566.2026.00021 · Cited by 4 · see citing works
Agent-Based Intelligent Operations and Maintenance for Artificial Intelligence Data Centers
Niu, Y., Li, X., Zhang, D., Han, B., Zhou, Y., & Xu, B. (2025) · IEEE International Conference on High Performance Computing and Communications; IEEE International Conference on Smart City; IEEE International Conference on Data Science and Systems (HPCC/SmartCity/DSS) · doi:10.1109/hpcc67675.2025.00221 · Cited by 1 · see citing works
LLMGuard: Multi-Agent Fault Diagnosis for Reliable Language-Model-as-a-Service
Zhong, Y., Yu, G., Huang, Y., Fu, Q., Ren, R., Feng, C., Yang, Y., & Lyu, M. R. (2026) · International Conference on Dependable Systems & Networks/Proceedings · doi:10.1109/dsn69566.2026.00018 · Cited by 1 · see citing works
When LLMs Listen to Experts: Accurate Failure Diagnosis in Operating Systems
Zhao, Y., Zhang, S., Sun, Y., Gu, W., Sun, Y., Wang, L., Shi, L., Huang, C., Yang, G., Zhang, L., et al. (2026) · IEEE/ACM International Conference on Software Engineering: Software Engineering in Practice (ICSE-SEIP) · doi:10.1145/3786583.3786852 · Cited by 1 · see citing works
Self-citationJunaid, E. (2026) · Research Square · doi:10.21203/rs.3.rs-11209235/v1 · Cited by 1 · see citing works
Wang, J., Chen, R., Wang, Y., Liao, X., Fan, Z., Wu, X., Wu, W., Liu, X., & Zhao, J. (2026) · 2026 IEEE International Conference on Web Services (ICWS) · doi:10.1109/icws72778.2026.00056
Large language models for agentic NetOps and AIOps: Architectures, evaluation, and safety
Bilal, M., Crowcroft, J., Wang, R., Xu, X., & Dustdar, S. (2027) · Computer Science Review · doi:10.1016/j.cosrev.2026.101075
Uncertainty-Aware Multimodal Anomaly Detection for Microservice Systems With Active Learning
Dou, H., Guo, S., Zou, L., Zhang, Y., Chen, P., & Zheng, Z. (2026) · IEEE Transactions on Services Computing · doi:10.1109/tsc.2026.3672587
Emerging Trends and Future Directions
Du, H., Chen, X., Liu, Y., & Huang, K. (2026) · Wireless networks · doi:10.1007/978-3-032-20194-2_7
Self-citationSTAR: A Stage-Attributed Triage and Repair Framework for RCA Agents in Microservices
Wang, J., Chen, R., Wang, Y., Liao, X., Fan, Z., Wu, X., Wu, W., Zhao, J., & Liu, X. (2026) · IEEE International Conference on Web Services (ICWS) · doi:10.1109/icws72778.2026.00057 · Cited by 1 · see citing works
CORE - An LLM-Powered System for Resolution Mining and Retrieval from Conversational Platforms
Dutta, A., Sisodiya, V. R., Airen, H., & Nivarthi, P. (2026) · IEEE International Conference on Data Engineering Workshops (ICDEW) · doi:10.1109/icdew71238.2026.00020
Toward In-Depth Root Cause Localization for Microservices With Multi-Agent Recursion-of-Thought
Zhang, L., Jia, T., Wang, K., Duan, C., He, M., Wang, R., Peng, X., Wang, M., Zhang, G., Chen, R., et al. (2026) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2026.3706572
SCALER: Self-Consistent Assessment for SLA Explanation and Resolution in 6G Networks
Panou, Z., Selis, D., Ramantas, K., Alonso, L., & Verikoukis, C. V. (2026) · International Conference on Communications · doi:10.1109/icc59461.2026.11588231
Li, X., Sun, N., & Jiang, J. (2026) · Lecture notes in computer science · doi:10.1007/978-981-92-3015-0_19
Multi-Agent System in the AIOps Paradigm: A Layered Architecture for System Telemetry Analysis
Pushkarev, D. A. (2026) · IEEE International Conference on Soft Computing and Measurements (SCM) · doi:10.1109/scm71573.2026.11625923
TeleCom-Bench: How Far Are Large Language Models from Industrial Telecommunication Applications?
Xiao, J., Lin, Y., Qiu, H., Ma, R., Zhong, C., Xu, D., Long, X., Hao, Q., Zou, D., Yang, Z., et al. (2026) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3770855.3817480
Ren, Y.-Y., Lin, X., Wang, C., Chen, L., Wang, S., Gao, K., Li, D., Bai, Y., Tian, C., Zhang, X., et al. (2026) · ACM SIGCOMM Conference (SIGCOMM) · doi:10.1145/3789240.3829145
SOPSmith: Forging Executable SOPs for LLM-Driven GPU Cluster Network Diagnosis
Yu, G., Sun, X., Huang, Z., Shi, Y., Xiao, X., Song, Y., Li, X., Lyu, B., Liu, Z., & He, Q. (2026) · International Workshop on Quality of Service · doi:10.1109/iwqos70441.2026.11661084
Cloud Microservices: An Empirical Study of MCP-Orchestrated LLM Agents for Incident Triage
Souza, R. G. D. S., & Júnior, M. C. (2026) · doi:10.5753/semish.2026.21828
RCAgentBench: An Agent-Oriented Benchmark for Multimodal Root Cause Analysis in Microservices
Jiang, H., Wang, Z., Nie, X., Gao, D., Li, J., & Pei, C. (2026) · International Workshop on Quality of Service · doi:10.1109/iwqos70441.2026.11661026
Self-citationSCALER: LLM-Based Cross-Modal Alignment for Microservice Root Cause Analysis
Yang, S., Wang, G., Ming, Z., He, D., Cui, L., Yu, X., & Zhou, X. (2026) · IEEE International Conference on Web Services (ICWS) · doi:10.1109/icws72778.2026.00101
ReAct Ops: A Plan-and-Execute ReAct Agent Framework for Autonomous Distributed Cluster Diagnosis
Li, D., Miao, Y., Chen, J., & Chen, L. (2026) · International Conference on Computer Information Science and Application Technology (CISAT) · doi:10.1109/cisat71030.2026.11682211
Self-citationFoundRoot: Towards Foundation Model for Root Cause Analysis via Structured Deep Thinking
Xie, Z., Li, Z., He, X., Zhang, S., Xu, L., Yang, Y., Zhang, T., Chen, J., Shi, R., & Pei, D. (2026) · ACM/IEEE International Conference on Software Engineering: Companion Proceedings (ICSE-Companion) · doi:10.1145/3744916.3787814
Self-citationClosed-Loop LLM-Driven Self-Healing for Enterprise Data Pipelines
Sivaraman, M. K. (2026) · Technology and Innovation Conference (TIC) · doi:10.1109/tic68483.2026.11703022
From Context to Code: Codifying Cross-System Reasoning for LLM-Assisted Operations
Chaudhry, T. (2026) · IEEE Access · doi:10.1109/access.2026.3738134
NSync: Automated Cloud Infrastructure-as-Code Reconciliation with AI Agents
Yang, Z., Guan, H., Nicolet, V., Paulsen, B., Dodds, J., Kroening, D., & Chen, A. (2026) · Proceedings of the ACM on software engineering. · doi:10.1145/3832121
LLM-Assisted Joint Ticket and Log Analysis for Incident Triage in Intelligent and Connected Vehicles
Fu, R., Zhang, S., Gu, W., Li, W., Sun, Y., & Pei, D. (2026) · IEEE/ACM International Conference on Automated Software Engineering (ASE) · doi:10.1145/3832783.3834521
Self-citationHERO: Hypothesis-Centered Root-Cause Analysis for Microservice Incidents
Lyu, J., Yi, J., Liang, S., Chen, P., & Pan, L. (2026) · IEEE/ACM International Conference on Automated Software Engineering (ASE) · doi:10.1145/3832783.3837509
DynRCA: Dynamic Sampling for Efficient Multi-Agent Reasoning in Microservice RCA
Liu, Y., Xiao, Y., Yang, G., Guo, S., & Qi, F. (2026) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn69854.2026.11728860
Self-citation
- Authors
- Pei, C., Wang, Z., Liu, F., Li, Z., Liu, Y., He, X., Kang, R., Zhang, T., Chen, J., Li, J., Xie, G., & Pei, D.
- Affiliations
- Chinese Academy of Sciences (CN)
- ByteDance (China) (CN)
- ByteDance, San Jose, USA
- Tsinghua University (CN)
- ORCID
- Venue
- Companion Proceedings of the ACM on Web Conference 2025
- Publisher
- ACM
- Published
- 2025-05-08
- Pages
- 422-431
- OpenAlex
- W4410636386
- Cited by
- 33
- References
- 24
- Citations by year
- 2026: 29 · 2025: 4
- FWCI
- 41.13
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- International World Wide Web Conference (WWW) (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Chinese Academy of Sciences (241711KYSB20200023)
- National Natural Science Foundation of China (62202445)
- National Natural Science Foundation of China-Research Grants Council (RGC) Joint Research Scheme (62321166652)
- Topics
- Physical Sciences › Computer Science › Computer Networks and Communications › Software System Performance and Reliability
- Physical Sciences › Computer Science › Information Systems › Software Engineering Research
- Physical Sciences › Computer Science › Software › Software Testing and Debugging Techniques
- Keywords
- root cause analysis, large language models, multi-agent systems, microservices, incident diagnosis, hallucination mitigation, ReAct framework, site reliability engineering
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
