ChronosAD: Leveraging Time Series Foundation Models for Accurate Anomaly Detection

The official implementation of the paper ChronosAD: Leveraging Time Series Foundation Models for Accurate Anomaly Detection accepted at the 24th IEEE International Conference on Industrial Informatics (INDIN 2026).

Abstract

Time series anomaly detection is a crucial task in various domains, including finance, healthcare, and industry. However, existing methods often struggle to generalize across different datasets, especially when anomalies are subtle or context-dependent. To solve this issue, we introduce ChronosAD, a novel architecture for anomaly detection that uses a time series foundation model as a feature extractor. Specifically, it employs a two-stage pipeline: first, it uses the foundation model to extract embeddings for each time series in a zero-shot manner. Then, a custom-developed Temporal Block, composed of Bidirectional Long Short-Term Memory (BiLSTM) and Multi-Head Attention, refines these embeddings to capture temporal dependencies and highlight salient patterns. Unlike previous approaches, our model requires minimal task-specific tuning and demonstrates robust generalization across a wide range of domains, including industrial, medical, cyber-physical, and automotive systems. Extensive experiments on 11 benchmarks show that ChronosAD outperforms existing methods by 4.72% in AUC and 6.60% in AP on average.

Installation

1. Repository setup:

$ git clone https://github.com/intelligolabs/ChronosAD.git
$ cd ChronosAD

2. Environment setup:

$ conda create -n ChronosAD python=3.10
$ conda activate ChronosAD
$ pip install git+https://github.com/amazon-science/chronos-forecasting.git
$ pip install -r requirements.txt

3. Training and Testing:

$ python main.py <dataset_path> --model ChronosLSTM --dataset_name <dataset_name> --batch_size 32 --context 250 --num_epochs 30 --learning_rate 0.001 --gpu_idx 0 --qualitative_results

Acknowledgement

We gratefully acknowledge the open-source contributions of the Chronos Foundation Model team, whose work made this project possible.

Authors

Uzair Khan¹, Luigi Capogrosso², Francesco Biondani¹, Michele Magno^3,2, Franco Fummi¹, Francesco Setti¹, Marco Cristani^1,4

¹ University of Verona ² Interdisciplinary Transformation University of Austria ³ ETH Zurich ⁴ Reykjavik University