mlsanityData to 7 Oct 2026

Log-based Anomaly Detection with Deep Learning: How Far Are We?

arXiv 2202.04301 ↗First posted 9 Feb 2022cs.SE cs.LGcited by 56

Not on the map. Why?

Abstract (arXiv)

Software-intensive systems produce logs for troubleshooting purposes. Recently, many deep learning models have been proposed to automatically detect system anomalies based on log data. These models typically claim very high detection accuracy. For example, most models report an F-measure greater than 0.9 on the commonly-used HDFS dataset. To achieve a profound understanding of how far we are from solving the problem of log-based anomaly detection, in this paper, we conduct an in-depth analysis of five state-of-the-art deep learning-based models for detecting system anomalies on four public log datasets. Our experiments focus on several aspects of model evaluation, including training data selection, data grouping, class distribution, data noise, and early detection ability. Our results point out that all these aspects have significant impact on the evaluation, and that all the studied models do not always work well. The problem of log-based anomaly detection has not been solved yet. Based on our findings, we also suggest possible future work.

Key facts. Cited by 56 papers in the map (data through 7 Oct 2026). Builds on 11 of its 51 references that are in the map. First posted on arXiv 9 Feb 2022.

Builds on

11 of 51 references.

References · 11

  1. 2017Attention Is All You Needcited by 59,932
  2. 2014Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translationcited by 4,016
  3. 2014Convolutional Neural Networks for Sentence Classificationcited by 1,124
  4. 2016FastText.zip: Compressing text classification modelscited by 331
  5. 2021LogBERT: Log Anomaly Detection via BERTcited by 114
  6. 2021Log-based Anomaly Detection Without Log Parsingcited by 59
  7. 2018Tools and Benchmarks for Automated Log Parsingcited by 56
  8. 2020A Survey on Automated Log Analysis for Reliability Engineeringcited by 50
  9. 2021Experience Report: Deep Learning-based System Log Analysis for Anomaly Detectioncited by 37
  10. 2020Self-Attentive Classification-Based Anomaly Detection in Unstructured Logscited by 36
  11. 2020Logram: Efficient Log Parsing Using n-Gram Dictionariescited by 9

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 5; 2023: 18; 2024: 11; 2025: 12; 2026: 10.

56 papers.

Sort:

By area:

  1. 2023RCAgent: Cloud Root Cause Analysis by Autonomous Agents with Tool-Augmented Large Language Modelscited by 62
  2. 2023Interpretable Online Log Analysis Using Large Language Models with Prompt Strategiescited by 34
  3. 2023LogGPT: Exploring ChatGPT for Log-Based Anomaly Detectioncited by 33
  4. 2022Deep Learning for Anomaly Detection in Log Data: A Surveycited by 27
  5. 2023LogGPT: Log Anomaly Detection via GPTcited by 26
  6. 2024LogLLM: Log-based Anomaly Detection Using Large Language Modelscited by 21
  7. 2022Tele-Knowledge Pre-training for Fault Analysiscited by 18
  8. 2023Log Parsing: How Far Can ChatGPT Go?cited by 16
  9. 2023AI for IT Operations (AIOps) on Cloud Platforms: Reviews, Opportunities and Challengescited by 13
  10. 2024LogLM: From Task-based to Instruction-based Automated Log Analysiscited by 13
  11. 2022LogGD:Detecting Anomalies from System Logs by Graph Neural Networkscited by 8
  12. 2023Graph Neural Networks based Log Anomaly Detection and Explanationcited by 8
38 more
  1. 2023A Critical Review of Common Log Data Sets Used for Evaluation of Sequence-based Anomaly Detection Techniquescited by 7
  2. 2024Reducing Events to Augment Log-based Anomaly Detection Models: An Empirical Studycited by 7
  3. 2023On the Effectiveness of Log Representation for Log-based Anomaly Detectioncited by 6
  4. 2023Twin Graph-based Anomaly Detection via Attentive Multi-Modal Learning for Microservice Systemcited by 6
  5. 2024Demystifying and Extracting Fault-indicating Information from Logs for Failure Diagnosiscited by 6
  6. 2025K4: Online Log Anomaly Detection Via Unsupervised Typicality Learningcited by 6
  7. 2024A Comparative Study on Large Language Models for Log Parsingcited by 5
  8. 2024LogELECTRA: Self-supervised Anomaly Detection for Unstructured Logscited by 4
  9. 2025AdaptiveLog: An Adaptive Log Analysis Framework with the Collaboration of Large and Small Language Modelcited by 4
  10. 2022Feature Selection for Fault Detection and Prediction based on Event Log Analysiscited by 3
  11. 2023Log-based Anomaly Detection based on EVT Theory with feedbackcited by 3
  12. 2023A Comprehensive Study of Machine Learning Techniques for Log-Based Anomaly Detectioncited by 3
  13. 2025Adapting Large Language Models for Parameter-Efficient Log Anomaly Detectioncited by 3
  14. 2023LogAI: A Library for Log Analytics and Intelligencecited by 2
  15. 2023Anomaly Detection of Command Shell Sessions based on DistilBERT: Unsupervised and Supervised Approachescited by 2
  16. 2024HELP: Hierarchical Embeddings-based Log Parsingcited by 2
  17. 2024LUK: Empowering Log Understanding with Expert Knowledge from Large Language Modelscited by 2
  18. 2025Data and System Perspectives of Sustainable Artificial Intelligencecited by 2
  19. 2025Hypergraph-Guided Regex Filter Synthesis for Event-Based Anomaly Detectioncited by 2
  20. 2023Try with Simpler -- An Evaluation of Improved Principal Component Analysis in Log-based Anomaly Detectioncited by 1
  21. 2023Log-based Anomaly Detection of Enterprise Software: An Empirical Studycited by 1
  22. 2025ADALog: Adaptive Unsupervised Anomaly detection in Logs with Self-attention Masked Language Modelcited by 1
  23. 2025A Novel GPT-Based Framework for Anomaly Detection in System Logscited by 1
  24. 2025FedLAD: A Modular and Adaptive Testbed for Federated Log Anomaly Detectioncited by 1
  25. 2026LLM-Enhanced Log Anomaly Detection: A Comprehensive Benchmark of Large Language Models for Automated System Diagnosticscited by 1
  26. 2022syslrn: Learning What to Monitor for Efficient Anomaly Detectioncited by 0
  27. 2023CSCLog: A Component Subsequence Correlation-Aware Log Anomaly Detection Methodcited by 0
  28. 2023RAPID: Training-free Retrieval-based Log Anomaly Detection with PLM considering Token-level informationcited by 0
  29. 2024Semi-supervised learning via DQN for log anomaly detectioncited by 0
  30. 2024MLAD: A Unified Model for Multi-system Log Anomaly Detectioncited by 0
  31. 2024What Information Contributes to Log-based Anomaly Detection? Insights from a Configurable Transformer-Based Approachcited by 0
  32. 2025Beyond Window-Based Detection: A Graph-Centric Framework for Discrete Log Anomaly Detectioncited by 0
  33. 2025RationAnomaly: Log Anomaly Detection with Rationality via Chain-of-Thought and Reinforcement Learningcited by 0
  34. 2025LogPurge: Log Data Purification for Anomaly Detection via Rule-Enhanced Filteringcited by 0
  35. 2025A unified framework for detecting point and collective anomalies in operating system logs via collaborative transformerscited by 0
  36. 2026Log anomaly detection via Meta Learning and Prototypical Networks for Cross domain generalizationcited by 0
  37. 2026ContraLog: Log File Anomaly Detection with Contrastive Learning and Masked Language Modelingcited by 0
  38. 2026KRONE: Scalable LLM-Augmented Log Anomaly Detection via Hierarchical Abstractioncited by 0

50 most cited of 56 All 56 (JSON)