Forecast2Anomaly (F2A): Adapting Multivariate Time Series Foundation Models for Anomaly Prediction
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911249094148096 |
|---|---|
| author | Hassan, Atif Kumar, Tarun Mishra, Ashish Serebryakov, Sergey Mopur, Satish Kumar Koganti, Phanidhar Chelankuri, Murthy Vogety, Ramanagopal Bhattacharya, Suparna Foltin, Martin |
| author_facet | Hassan, Atif Kumar, Tarun Mishra, Ashish Serebryakov, Sergey Mopur, Satish Kumar Koganti, Phanidhar Chelankuri, Murthy Vogety, Ramanagopal Bhattacharya, Suparna Foltin, Martin |
| contents | Forecasting anomalies (anomaly prediction) in multivariate time series from different real-world, dynamic, and complex systems is vital for preempting critical failures, leading to a substantial minimization in operational costs and human labor. Yet, existing methods are limited to specific systems while failing to generalize to evolving anomaly patterns over time. In contrast, pretrained Time Series Foundation Models (TSFMs) have recently demonstrated strong generalization and zero-shot forecasting capabilities. However, their potential remains untapped for anomaly prediction, a task fundamentally different from forecasting normal behavior. Thus, we present Forecast2Anomaly (F2A), a novel framework that empowers TSFMs with anomaly prediction abilities through two key innovations. First, we propose a joint forecast-anomaly loss that fine-tunes TSFMs to accurately forecast future signals even at anomalous time points. Second, we introduce a Retrieval-Augmented Generation (RAG) module that retrieves historically relevant horizons and conditions predictions on them. This component dynamically adapts to distributional shifts at inference time, enabling F2A to track evolving anomalies without requiring model updates. By combining targeted fine-tuning with dynamic retrieval, F2A bridges the gap between robust TSFM zero-shot forecasting and zero-shot anomaly prediction. Extensive experiments across 16 diverse datasets and multiple TSFM backbones show that F2A consistently outperforms state-of-the-art methods, offering a scalable, zero-shot anomaly prediction solution for real-world applications. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_03149 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Forecast2Anomaly (F2A): Adapting Multivariate Time Series Foundation Models for Anomaly Prediction Hassan, Atif Kumar, Tarun Mishra, Ashish Serebryakov, Sergey Mopur, Satish Kumar Koganti, Phanidhar Chelankuri, Murthy Vogety, Ramanagopal Bhattacharya, Suparna Foltin, Martin Machine Learning Artificial Intelligence Forecasting anomalies (anomaly prediction) in multivariate time series from different real-world, dynamic, and complex systems is vital for preempting critical failures, leading to a substantial minimization in operational costs and human labor. Yet, existing methods are limited to specific systems while failing to generalize to evolving anomaly patterns over time. In contrast, pretrained Time Series Foundation Models (TSFMs) have recently demonstrated strong generalization and zero-shot forecasting capabilities. However, their potential remains untapped for anomaly prediction, a task fundamentally different from forecasting normal behavior. Thus, we present Forecast2Anomaly (F2A), a novel framework that empowers TSFMs with anomaly prediction abilities through two key innovations. First, we propose a joint forecast-anomaly loss that fine-tunes TSFMs to accurately forecast future signals even at anomalous time points. Second, we introduce a Retrieval-Augmented Generation (RAG) module that retrieves historically relevant horizons and conditions predictions on them. This component dynamically adapts to distributional shifts at inference time, enabling F2A to track evolving anomalies without requiring model updates. By combining targeted fine-tuning with dynamic retrieval, F2A bridges the gap between robust TSFM zero-shot forecasting and zero-shot anomaly prediction. Extensive experiments across 16 diverse datasets and multiple TSFM backbones show that F2A consistently outperforms state-of-the-art methods, offering a scalable, zero-shot anomaly prediction solution for real-world applications. |
| title | Forecast2Anomaly (F2A): Adapting Multivariate Time Series Foundation Models for Anomaly Prediction |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2511.03149 |