RCAgent: Cloud Root Cause Analysis by Autonomous Agents with Tool-Augmented Large Language Models
Open accessThe DOI matches its Crossref record; 66 of the 66 references it cites were found in a registry or an index; 2 of the 26 sources its claims cite were checked for support.
Abstract
Large language model (LLM) applications in cloud root cause analysis (RCA) have been actively explored recently. However, current methods are still reliant on manual workflow settings and do not unleash LLMs' decision-making and environment interaction capabilities. We present RCAgent, a tool-augmented LLM autonomous agent framework for practical and privacy-aware industrial RCA usage. Running on an internally deployed model rather than GPT families, RCAgent is capable of free-form data collection and comprehensive analysis with tools. Our framework combines a variety of enhancements, including a unique Self-Consistency for action trajectories, and a suite of methods for context management, stabilization, and importing domain knowledge. Our experiments show RCAgent's evident and consistent superiority over ReAct across all aspects of RCA--predicting root causes, solutions, evidence, and responsibilities--and tasks covered or uncovered by current rules, as validated by both automated metrics and human evaluations. Furthermore, RCAgent has already been integrated into the diagnosis and issue discovery workflow of the Real-time Compute Platform for Apache Flink of Alibaba Cloud.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
1 Valid · 3 Not valid · 15 Not checked · 67 Own results and interpretations
None of the sources cited for this claim have been checked yet.
In the paper
“However, the adoption of LLM agents in the AIOps field, especially with noisy and lengthy data, remains limited [21,28].”
§ RCAgent: Cloud Root Cause Analysis by Autonomous Agents with Tool-Augmented Large Language Models > 1 INTRODUCTION
What the sources say
- Could not verifycites [21] matched to “Log-based anomaly detection with deep learning: how far are we?” (2022) sourceExperimental — accuracy not yet measured
Our copy of Le & Zhang (ICSE 2022) is an empirical study of five pre-LLM deep learning log-anomaly-detection models (DeepLog, LogAnomaly, PLELog, LogRobust, CNN; Sections 2.1 and 3.2, pp. 2-4) and contains no discussion of LLMs, LLM agents, or their adoption in AIOps — searched: 'large language model/LLM/GPT/agent', 'AIOps', 'adoption', with no passage on the point. The closest material is on data noise: Section 3.1.4 (p. 4) on mislabelled logs and log-parsing errors, and Section 4.4 (pp. 8-9) showing F-measures collapse with 1% mislabelling and vary across parsers — but that concerns conventional DL models, not LLM agents, so it neither supports nor contradicts the attributed claim. The claim's substance therefore cannot be checked against this source. (LLM, conf=0.75)
- Not checked yetcites [28] matched to “LogAssist: Assisting Log Analysis Through Log Summarization” (2022) source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Could not verify
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“However, these models are not aware of the workflow of cloud RCA, leaving them simply analytical tools.”
§ RCAgent: Cloud Root Cause Analysis by Autonomous Agents with Tool-Augmented Large Language Models > 7.3 Cloud RCA with LLMs
What the sources say
- Not checked yetcites [2]
- Not checked yetcites [17]
- Partly supportedcites [10] matched to “Automatic Root Cause Analysis via Large Language Models for Cloud Incidents” (2024) sourceExperimental — accuracy not yet measured
Our copy does critique existing LLM-based cloud RCA and positions RCACopilot as the workflow-aware alternative (Introduction, p.1: Ahmed et al. fine-tune an LLM 'just by leveraging the title and summary'; limitations listed as narrow input signals, costly finetuning, difficulty updating, and increased hallucination over time; Related Work, p.7: 'Different from prior work, we build a system that can automatically integrate metrics, logs, and traces'; Introduction: RCACopilot's 'incident handlers—automated workflows tailored to each alert type'). But the specific characterization attributed here is absent: nowhere does the paper say these models are 'not aware of the workflow of cloud RCA' or that they 'remain simply analytical tools' — its stated limitations are about data scope, cost, updatability and hallucination, and it in fact describes the prior LLM work (Ahmed et al.) as recommending 'root causes and mitigation steps' (Related Work, p.7), i.e. not purely analytical. The workflow-unawareness framing and the 'simply analytical tools' conclusion are the citing authors'. (LLM, conf=0.68)
- Not checked yetcites [16]
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Employing TSC aggregation using LLM summarization, the overall performance of RCAgent gains further enhancements, especially on solution prediction, witnessing gains of +3.51 METEOR, +4.50 BLEURT, and +2.28% G-Helpfulness.”
§ RCAgent: Cloud Root Cause Analysis by Autonomous Agents with Tool-Augmented Large Language Models > 5 RESULT 5.1 Effectiveness
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The analysis results from RCAgent are being utilized in the Real-time Compute Platform for Apache Flink of Alibaba Cloud to diagnose anomalous stream processing jobs uncovered by current methods.”
§ RCAgent: Cloud Root Cause Analysis by Autonomous Agents with Tool-Augmented Large Language Models > 1 INTRODUCTION
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“We demonstrate the practical usage of RCAgent with realworld experiments on computing jobs in Alibaba Cloud.”
§ RCAgent: Cloud Root Cause Analysis by Autonomous Agents with Tool-Augmented Large Language Models > 1 INTRODUCTION
boosted- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
26 more claims
Create a free account to see all 86 claims and to verify the 24 references still open.
- It may state two separate claims as one. 36
- Probably not a substantial claim. 25
- The quote only partly states this claim. 2
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?
The cited source says what the claim attributes to it.
In the citing paper
“RCAgent [9] is an autonomous, tool-augmented LLM agent for practical cloud RCA, which iteratively collects evidence from observability tools and synthesizes a diagnosis.”
From TopoEvo: A Topology-Aware Self-Evolving Multi-Agent Framework for Root Cause Analysis in Microservices (2026) · cites this paper as [9]
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“For confidence gating and abstention, calibration matters, because wrong-but-confident agents create the worst operational incidents [104], [141].”
From Large language models for agentic NetOps and AIOps: Architectures, evaluation, and safety (2027) · cites this paper as [104]
Experimental — accuracy not yet measured
Citations
Cites
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
Fast unfolding of communities in large networks
Blondel, V. D., Guillaume, J.-L., Lambiotte, R., & Lefebvre, E. (2008) · Journal of Statistical Mechanics: Theory and Experiment · doi:10.1088/1742-5468/2008/10/p10008 · Cited by 21,424
LLaMA: Open and Efficient Foundation Language Models
Touvron, H., Lavril, T., Izacard, G., Martinet, X., Lachaux, M.-A., Lacroix, T., Rozière, B., Goyal, N., Hambro, E., Azhar, F., et al. (2023) · arXiv · doi:10.48550/arxiv.2302.13971 · Cited by 4,602
Chain-Of-Thought Prompting Elicits Reasoning in Large Language Models
Wei, J., Wang, X., Schuurmans, D., Bosma, M., Ichter, B., Xia, F., Chi, E., Le, Q. V., & Zhou, D. (2022) · Advances in Neural Information Processing Systems 35 · doi:10.52202/068431-1800 · Cited by 4,505
Training language models to follow instructions with human feedback
Ouyang, L., Wu, J., Jiang, X., Almeida, D., Wainwright, C. L., Mishkin, P., Zhang, C., Agarwal, S., Slama, K., Ray, A., et al. (2022) · arXiv · doi:10.48550/arxiv.2203.02155 · Cited by 4,208
Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
Lewis, P., Perez, E., Piktus, A., Petroni, F., Karpukhin, V., Goyal, N., Küttler, H., Lewis, M., Yih, W.-T., Rocktäschel, T., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.11401 · Cited by 3,101
Llama 2: Open Foundation and Fine-Tuned Chat Models
Touvron, H., Martin, L., Stone, K., Albert, P., Almahairi, A., Babaei, Y., Bashlykov, N., Batra, S., Bhargava, P., Bhosale, S., et al. (2023) · arXiv · doi:10.48550/arxiv.2307.09288 · Cited by 2,604 · cited 1×
“Theembedding modelweuseis GTE-LARGE [24],foritsslightly”
Self-citationGenerative Agents: Interactive Simulacra of Human Behavior
Park, J. S., O'Brien, J. C., Cai, C. J., Morris, M. R., Liang, P., & Bernstein, M. S. (2023) · arXiv · doi:10.48550/arxiv.2304.03442 · Cited by 2,173
Large language models are zero-shot reasoners
Kojima, T., Gu, S. (., Reid, M., Matsuo, Y., & Iwasawa, Y. (2022) · Advances in Neural Information Processing Systems 35 · doi:10.52202/068431-1613 · Cited by 2,013
Large Language Models are Zero-Shot Reasoners
Kojima, T., Gu, S. S., Reid, M., Matsuo, Y., & Iwasawa, Y. (2022) · arXiv · doi:10.48550/arxiv.2205.11916 · Cited by 2,013
Research works printed but not linked: 6
- Pooja Aggarwal, Ajay Gupta, Prateeti Mohapatra, Seema Nagar, Atri Mandal, Qing Wang, and Amit Paradkar. 2020. Localization of operational faults in cloud applications by mining causal dependencies in logs using golden signals. In International Conference on Service-Oriented Computing. 137-149.
- Satanjeev Banerjee and Alon Lavie. 2005. METEOR: An automatic metric for MT evaluation with improved correlation with human judgments. In ACL workshop on Intrinsic and Extrinsic Evaluation Measures for Machine Translation and/or Summarization. 65-72.
- Junjie Chen, Xiaoting He, Qingwei Lin, Yong Xu, Hongyu Zhang, Dan Hao, Feng Gao, Zhangwei Xu, Yingnong Dang, and Dongmei Zhang. 2019. An empiri- cal investigation of incident triage for online service systems. In International Conference on Software Engineering: Software Engineering in Practice. 111-120.
- Supriyo Ghosh, Manish Shetty, Chetan Bansal, and Suman Nath. 2022. How to fight production incidents? an empirical study on a large-scale cloud service. In Symposium on Cloud Computing. 126-141.
- Woosuk Kwon, Zhuohan Li, Siyuan Zhuang, Ying Sheng, Lianmin Zheng, Cody Hao Yu, Joseph Gonzalez, Hao Zhang, and Ion Stoica. 2023. Efficient Memory Management for Large Language Model Serving with PagedAttention. In Symposium on Operating Systems Principles. 611-626.
- Xuezhi Wang, Jason Wei, Dale Schuurmans, Quoc V Le, Ed H. Chi, Sharan Narang, Aakanksha Chowdhery, and Denny Zhou. 2023. Self-Consistency Improves Chain of Thought Reasoning in Language Models. In International Conference on Learning Representations. 1-24.
Cited by
A Survey on the Memory Mechanism of Large Language Model based Agents
Zhang, Z., Bo, X., Ma, C., Li, R., Chen, X., Dai, Q., Zhu, J., Dong, Z., & Wen, J.-R. (2024) · arXiv · doi:10.48550/arxiv.2404.13501 · Cited by 149
Large Language Models for Zero Touch Network Configuration Management
Lira, O. G., Rendón, O. M. C., & da Fonseca, N. L. S. (2024) · IEEE Communications Magazine · doi:10.1109/mcom.001.2400368 · Cited by 95
“RCAgent [7] proposes a practical and privacy-aware industrial root cause analysis solution.”
BackgroundLiu, Y., Lo, S. K., Lu, Q., Zhu, L., Zhao, D., Xu, X., Harrer, S., & Whittle, J. (2025) · Journal of Systems and Software · doi:10.1016/j.jss.2024.112278 · Cited by 51
From language to action: a review of large language models as autonomous agents and tool users
Chowa, S. S., Alvi, R., Rahman, S. S., Rahman, M. A., Raiaan, M. A. K., Islam, M. R., Hussain, M., & Azam, S. (2026) · Artificial Intelligence Review · doi:10.1007/s10462-025-11471-9 · Cited by 50
“Conversation-centric agents often adopt Vicuna-13B or the larger Wizard-Vicuna-Uncensored 30B [52,87,94,106,124,129,130], while the instruction-tuned WizardLM 70B serves as a high-capacity but fully open chat baseline [69].”
InfluentialWu, T., Li, J., Bao, J., & Liu, Q. (2024) · Journal of Manufacturing Systems · doi:10.1016/j.jmsy.2024.08.008 · Cited by 45
Metadata
- Authors
- Wang, Z., Liu, Z., Zhang, Y., Zhong, A., Wang, J., Yin, F., Fan, L., Wu, L., & Wen, Q.
- Affiliations
- Tsinghua University (CN)
- Nanjing University (CN)
- Alibaba Group (China) (CN)
- Harvard University (US)
- Xi'an Jiaotong University (CN)
- Anytime AI, New York, USA
- Squirrel Ai Learning, Bellevue, USA
- ORCID
- Venue
- Proceedings of the 33rd ACM International Conference on Information and Knowledge Management
- Publisher
- ACM
- Published
- 2024-10-21
- Pages
- 4966-4974
- ISSN
- 2155-0751 (ISSN-L)
- OpenAlex
- W4403582420
- Semantic Scholar
- 80a5aec9d2a972a1c4b4ead082ff256dc1115efb
- Cited by
- 99
- References
- 17
- Citations by year
- 2026: 63 · 2025: 30 · 2024: 6
- FWCI
- 43.34
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- ACM International Conference on Information and Knowledge Management (CIKM) (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Computer Networks and Communications › Software System Performance and Reliability
- Physical Sciences › Computer Science › Information Systems › Cloud Computing and Resource Management
- Physical Sciences › Computer Science › Information Systems › Software Engineering Research
- Keywords
- autonomous agents, self-consistency, context management, Apache Flink
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-07, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-07 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (qwen3.8-omni-flash).
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: RCAgent: Cloud Root Cause Analysis by Autonomous Agents with Tool-Augmented Large Language Models. https://app.cochat.ai/papers/10.1145/3627673.3680016
- The paper
- Wang, Z., Liu, Z., Zhang, Y., Zhong, A., Wang, J., Yin, F., Fan, L., Wu, L., & Wen, Q. (2024). RCAgent: Cloud Root Cause Analysis by Autonomous Agents with Tool-Augmented Large Language Models. Proceedings of the 33rd ACM International Conference on Information and Knowledge Management. https://doi.org/10.1145/3627673.3680016
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.