# MADLLM: Multivariate Anomaly Detection via Pre-trained LLMs

arXiv: https://arxiv.org/abs/2504.09504 · HTML: https://mlsanity.com/paper/2504.09504/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 21 references that are in the map.
- First posted on arXiv 13 Apr 2025.

## Builds on (most cited first, 13 of 13)

- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling](https://mlsanity.com/paper/1803.01271/) (2018)
- [TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis](https://mlsanity.com/paper/2210.02186/) (2022)
- [One Fits All:Power General Time Series Analysis by Pretrained LM](https://mlsanity.com/paper/2302.11939/) (2023)
- [Detecting Spacecraft Anomalies Using LSTMs and Nonparametric Dynamic Thresholding](https://mlsanity.com/paper/1802.04431/) (2018)
- [Unsupervised Scalable Representation Learning for Multivariate Time Series](https://mlsanity.com/paper/1901.10738/) (2019)
- [TranAD: Deep Transformer Networks for Anomaly Detection in Multivariate Time Series Data](https://mlsanity.com/paper/2201.07284/) (2022)
- [TEST: Text Prototype Aligned Embedding to Activate LLM's Ability for Time Series](https://mlsanity.com/paper/2308.08241/) (2023)
- [ImDiffusion: Imputed Diffusion Models for Multivariate Time Series Anomaly Detection](https://mlsanity.com/paper/2307.00754/) (2023)
- [Large Language Models can Deliver Accurate and Interpretable Time Series Anomaly Detection](https://mlsanity.com/paper/2405.15370/) (2024)
- [Large Language Model Guided Knowledge Distillation for Time Series Anomaly Detection](https://mlsanity.com/paper/2401.15123/) (2024)
- [Cocktail: Chunk-Adaptive Mixed-Precision Quantization for Long-Context LLM Inference](https://mlsanity.com/paper/2503.23294/) (2025)
- [INCPrompt: Task-Aware incremental Prompting for Rehearsal-Free Class-incremental Learning](https://mlsanity.com/paper/2401.11667/) (2024)

## Built on by (most cited first, 3 of 3)

- [MoQAE: Mixed-Precision Quantization for Long-Context LLM Inference via Mixture of Quantization-Aware Experts](https://mlsanity.com/paper/2506.07533/) (2025)
- [Vista: Scene-Aware Optimization for Streaming Video Question Answering under Post-Hoc Queries](https://mlsanity.com/paper/2602.08448/) (2026)
- [ASTER: Latent Pseudo-Anomaly Generation for Unsupervised Time-Series Anomaly Detection](https://mlsanity.com/paper/2604.13924/) (2026)

## Abstract (arXiv)

When applying pre-trained large language models (LLMs) to address anomaly detection tasks, the multivariate time series (MTS) modality of anomaly detection does not align with the text modality of LLMs. Existing methods simply transform the MTS data into multiple univariate time series sequences, which can cause many problems. This paper introduces MADLLM, a novel multivariate anomaly detection method via pre-trained LLMs. We design a new triple encoding technique to align the MTS modality with the text modality of LLMs. Specifically, this technique integrates the traditional patch embedding method with two novel embedding approaches: Skip Embedding, which alters the order of patch processing in traditional methods to help LLMs retain knowledge of previous features, and Feature Embedding, which leverages contrastive learning to allow the model to better understand the correlations between different features. Experimental results demonstrate that our method outperforms state-of-the-art methods in various public anomaly detection datasets.