When Will It Fail?: Anomaly to Prompt for Forecasting Future Anomalies in Time Series

Min-Yeong Park, Won-Jeong Lee, Seong Tae Kim, Gyeong-Moon Park
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:48086-48103, 2025.

Abstract

Recently, forecasting future abnormal events has emerged as an important scenario to tackle realworld necessities. However, the solution of predicting specific future time points when anomalies will occur, known as Anomaly Prediction (AP), remains under-explored. Existing methods dealing with time series data fail in AP, focusing only on immediate anomalies or failing to provide precise predictions for future anomalies. To address AP, we propose a novel framework called Anomaly to Prompt (A2P), comprised of Anomaly-Aware Forecasting (AAF) and Synthetic Anomaly Prompting (SAP). To enable the forecasting model to forecast abnormal time points, we adopt a strategy to learn the relationships of anomalies. For the robust detection of anomalies, our proposed SAP introduces a learnable Anomaly Prompt Pool (APP) that simulates diverse anomaly patterns using signal-adaptive prompt. Comprehensive experiments on multiple real-world datasets demonstrate the superiority of A2P over state-of-the-art methods, showcasing its ability to predict future anomalies.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-park25e, title = {When Will It Fail?: Anomaly to Prompt for Forecasting Future Anomalies in Time Series}, author = {Park, Min-Yeong and Lee, Won-Jeong and Kim, Seong Tae and Park, Gyeong-Moon}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {48086--48103}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/park25e/park25e.pdf}, url = {https://proceedings.mlr.press/v267/park25e.html}, abstract = {Recently, forecasting future abnormal events has emerged as an important scenario to tackle realworld necessities. However, the solution of predicting specific future time points when anomalies will occur, known as Anomaly Prediction (AP), remains under-explored. Existing methods dealing with time series data fail in AP, focusing only on immediate anomalies or failing to provide precise predictions for future anomalies. To address AP, we propose a novel framework called Anomaly to Prompt (A2P), comprised of Anomaly-Aware Forecasting (AAF) and Synthetic Anomaly Prompting (SAP). To enable the forecasting model to forecast abnormal time points, we adopt a strategy to learn the relationships of anomalies. For the robust detection of anomalies, our proposed SAP introduces a learnable Anomaly Prompt Pool (APP) that simulates diverse anomaly patterns using signal-adaptive prompt. Comprehensive experiments on multiple real-world datasets demonstrate the superiority of A2P over state-of-the-art methods, showcasing its ability to predict future anomalies.} }
Endnote
%0 Conference Paper %T When Will It Fail?: Anomaly to Prompt for Forecasting Future Anomalies in Time Series %A Min-Yeong Park %A Won-Jeong Lee %A Seong Tae Kim %A Gyeong-Moon Park %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-park25e %I PMLR %P 48086--48103 %U https://proceedings.mlr.press/v267/park25e.html %V 267 %X Recently, forecasting future abnormal events has emerged as an important scenario to tackle realworld necessities. However, the solution of predicting specific future time points when anomalies will occur, known as Anomaly Prediction (AP), remains under-explored. Existing methods dealing with time series data fail in AP, focusing only on immediate anomalies or failing to provide precise predictions for future anomalies. To address AP, we propose a novel framework called Anomaly to Prompt (A2P), comprised of Anomaly-Aware Forecasting (AAF) and Synthetic Anomaly Prompting (SAP). To enable the forecasting model to forecast abnormal time points, we adopt a strategy to learn the relationships of anomalies. For the robust detection of anomalies, our proposed SAP introduces a learnable Anomaly Prompt Pool (APP) that simulates diverse anomaly patterns using signal-adaptive prompt. Comprehensive experiments on multiple real-world datasets demonstrate the superiority of A2P over state-of-the-art methods, showcasing its ability to predict future anomalies.
APA
Park, M., Lee, W., Kim, S.T. & Park, G.. (2025). When Will It Fail?: Anomaly to Prompt for Forecasting Future Anomalies in Time Series. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:48086-48103 Available from https://proceedings.mlr.press/v267/park25e.html.

Related Material