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

arXiv: https://arxiv.org/abs/2605.05725 · HTML: https://mlsanity.com/paper/2605.05725/

## Key facts

- Cited by 1 papers in the map (data through 14 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.

## Placed in

- LLM agents and tool use / [Software engineering coding agents](https://mlsanity.com/agents/software-engineering-coding-agents/)

## Builds on (most cited first, 21 of 21)

- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Toolformer: Language Models Can Teach Themselves to Use Tools](https://mlsanity.com/paper/2302.04761/) (2023)
- [MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework](https://mlsanity.com/paper/2308.00352/) (2023)
- [CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society](https://mlsanity.com/paper/2303.17760/) (2023)
- [Time-LLM: Time Series Forecasting by Reprogramming Large Language Models](https://mlsanity.com/paper/2310.01728/) (2023)
- [Large Language Models Are Zero-Shot Time Series Forecasters](https://mlsanity.com/paper/2310.07820/) (2023)
- [Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancy](https://mlsanity.com/paper/2110.02642/) (2021)
- [LSTM-based Encoder-Decoder for Multi-sensor Anomaly Detection](https://mlsanity.com/paper/1607.00148/) (2016)
- [A Multimodal Anomaly Detector for Robot-Assisted Feeding Using an LSTM-based Variational Autoencoder](https://mlsanity.com/paper/1711.00614/) (2017)
- [PromptCast: A New Prompt-based Learning Paradigm for Time Series Forecasting](https://mlsanity.com/paper/2210.08964/) (2022)
- [Time-Series Anomaly Detection Service at Microsoft](https://mlsanity.com/paper/1906.03821/) (2019)
- [Imaging Time-Series to Improve Classification and Imputation](https://mlsanity.com/paper/1506.00327/) (2015)
- [Local Evaluation of Time Series Anomaly Detection Algorithms](https://mlsanity.com/paper/2206.13167/) (2022)
- [MALT: Improving Reasoning with Multi-Agent LLM Training](https://mlsanity.com/paper/2412.01928/) (2024)
- [TFAD: A Decomposition Time Series Anomaly Detection Architecture with Time-Frequency Analysis](https://mlsanity.com/paper/2210.09693/) (2022)
- [Large language models can be zero-shot anomaly detectors for time series?](https://mlsanity.com/paper/2405.14755/) (2024)
- [Large Language Models can Deliver Accurate and Interpretable Time Series Anomaly Detection](https://mlsanity.com/paper/2405.15370/) (2024)
- [Can LLMs Serve As Time Series Anomaly Detectors?](https://mlsanity.com/paper/2408.03475/) (2024)
- [Constructing Large-Scale Real-World Benchmark Datasets for AIOps](https://mlsanity.com/paper/2208.03938/) (2022)
- [See it, Think it, Sorted: Large Multimodal Models are Few-shot Time Series Anomaly Analyzers](https://mlsanity.com/paper/2411.02465/) (2024)
- [Can Multimodal LLMs Perform Time Series Anomaly Detection?](https://mlsanity.com/paper/2502.17812/) (2025)

## Built on by (most cited first, 1 of 1)

- [LLM Agents for Time-Series: A Survey](https://mlsanity.com/paper/2608.26226/) (2026)

## 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.