mlsanityData to 7 Oct 2026

Detecting Time Series Anomalies Like an Expert: A Multi-Agent LLM Framework with Specialized Analyzers

arXiv 2605.05725 ↗First posted 7 May 2026cs.AIcited by 1

On the map: LLM agents and tool use → Software engineering coding agents

Abstract (arXiv)

Time-series anomaly detection often returns scores or intervals, while analysts need to understand the abnormal behavior and the evidence supporting it. We introduce SAGE (Specialized Analyzer Group for Expert-like Detection), a multi-agent framework for evidence-grounded diagnosis of univariate time series. Four specialized Analyzers examine point, structural, seasonal, and pattern anomalies using numerical tools and diagnostic visualizations. A Detector integrates their evidence into intervals, candidate types, and evidence-strength confidence scores; a Supervisor translates these records into analyst-facing reports. Synthetic in-context references are constructed from normal-reference training segments, reducing dependence on real anomalous demonstrations. Across Yahoo S5, KPI, and WSD, SAGE achieves an average Point-F1 of 66.26, the highest among the evaluated methods. Controlled synthetic evaluation examines localization and type diagnosis, while component ablations support the detection contribution of specialized evidence generation. In a method-blind human study, evaluators rate SAGE's diagnostic outputs as more useful than those of the compared methods.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use. Builds on 21 of its 38 references that are in the map. First posted on arXiv 7 May 2026.

Builds on

21 of 38 references.

References · 21

  1. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,355
  2. 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,072
  3. 2023MetaGPT: Meta Programming for A Multi-Agent Collaborative Frameworkcited by 1,761
  4. 2023CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Societycited by 1,348
  5. 2023Time-LLM: Time Series Forecasting by Reprogramming Large Language Modelscited by 680
  6. 2023Large Language Models Are Zero-Shot Time Series Forecasterscited by 491
  7. 2021Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancycited by 387
  8. 2016LSTM-based Encoder-Decoder for Multi-sensor Anomaly Detectioncited by 213
  9. 2017A Multimodal Anomaly Detector for Robot-Assisted Feeding Using an LSTM-based Variational Autoencodercited by 209
  10. 2022PromptCast: A New Prompt-based Learning Paradigm for Time Series Forecastingcited by 170
  11. 2019Time-Series Anomaly Detection Service at Microsoftcited by 163
  12. 2015Imaging Time-Series to Improve Classification and Imputationcited by 85
9 more
  1. 2022Local Evaluation of Time Series Anomaly Detection Algorithmscited by 79
  2. 2024MALT: Improving Reasoning with Multi-Agent LLM Trainingcited by 71
  3. 2022TFAD: A Decomposition Time Series Anomaly Detection Architecture with Time-Frequency Analysiscited by 48
  4. 2024Large language models can be zero-shot anomaly detectors for time series?cited by 30
  5. 2024Large Language Models can Deliver Accurate and Interpretable Time Series Anomaly Detectioncited by 26
  6. 2024Can LLMs Serve As Time Series Anomaly Detectors?cited by 22
  7. 2024See it, Think it, Sorted: Large Multimodal Models are Few-shot Time Series Anomaly Analyzerscited by 20
  8. 2022Constructing Large-Scale Real-World Benchmark Datasets for AIOpscited by 19
  9. 2025Can Multimodal LLMs Perform Time Series Anomaly Detection?cited by 11

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026LLM Agents for Time-Series: A Surveycited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2026AnomaMind: Agentic Time Series Anomaly Detection with Tool-Augmented Reasoning
  2. 2025AXIS: Explainable Time Series Anomaly Detection with Large Language Models
  3. 2026AnomSeer: Reinforcing Multimodal LLMs to Reason for Time-Series Anomaly Detection
  4. 2025Towards Foundation Models for Zero-Shot Time Series Anomaly Detection: Leveraging Synthetic Data and Relative Context Discrepancy
  5. 2025Refining Time Series Anomaly Detectors using Large Language Models
  6. 2026ChatAD: Reasoning-Enhanced Time-Series Anomaly Detection with Multi-Turn Instruction Evolution
  7. 2024Can LLMs Understand Time Series Anomalies?
  8. 2026TraceBench: Controlled Evaluation of LLM Agents for Time-Series Root-Cause Attribution