xTime: Extreme Event Prediction with Hierarchical Knowledge Distillation and Expert Fusion

• URL: http://arxiv.org/abs/2510.20651v1
• Date: Thu, 23 Oct 2025 15:24:45 GMT
• Title: xTime: Extreme Event Prediction with Hierarchical Knowledge Distillation and Expert Fusion
• Authors: Quan Li, Wenchao Yu, Suhang Wang, Minhua Lin, Lingwei Chen, Wei Cheng, Haifeng Chen,
• Abstract summary: We propose xTime, a novel framework for extreme event forecasting in time series.<n>xTime leverages knowledge distillation to transfer information from models trained on lower-rarity events.<n>We introduce a mixture of experts (MoE) mechanism that dynamically selects and fuses outputs from expert models across different rarity levels.
• Score: 65.63135031712153
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Extreme events frequently occur in real-world time series and often carry significant practical implications. In domains such as climate and healthcare, these events, such as floods, heatwaves, or acute medical episodes, can lead to serious consequences. Accurate forecasting of such events is therefore of substantial importance. Most existing time series forecasting models are optimized for overall performance within the prediction window, but often struggle to accurately predict extreme events, such as high temperatures or heart rate spikes. The main challenges are data imbalance and the neglect of valuable information contained in intermediate events that precede extreme events. In this paper, we propose xTime, a novel framework for extreme event forecasting in time series. xTime leverages knowledge distillation to transfer information from models trained on lower-rarity events, thereby improving prediction performance on rarer ones. In addition, we introduce a mixture of experts (MoE) mechanism that dynamically selects and fuses outputs from expert models across different rarity levels, which further improves the forecasting performance for extreme events. Experiments on multiple datasets show that xTime achieves consistent improvements, with forecasting accuracy on extreme events improving from 3% to 78%.

Related papers

• Extreme-value forest fire prediction A study of the Loss Function in an Ordinality Scheme [2.2049183478692593]
  We introduce the first ordinal classification framework for forecasting wildfire severity levels directly aligned with operational decision-making in France.<n>Our study investigates the influence of loss-function design on the ability of neural models to predict rare yet critical high-severity fire occurrences.
  arXiv  Detail & Related papers  (2026-01-06T15:46:56Z)
• EventTSF: Event-Aware Non-Stationary Time Series Forecasting [73.54313384419792]
  EventTSF is an autoregressive generation framework that integrates historical time series with textual events to make subsequent forecasts.<n>Experiments on 8 synthetic and real-world datasets show that EventTSF outperforms 12 baselines across diverse event-aware non-stationary time series forecasting scenarios.
  arXiv  Detail & Related papers  (2025-08-19T01:28:47Z)
• OneForecast: A Universal Framework for Global and Regional Weather Forecasting [67.61381313555091]
  We propose a global-regional nested weather forecasting framework (OneForecast) based on graph neural networks.<n>By combining a dynamic system perspective with multi-grid theory, we construct a multi-scale graph structure and densify the target region.<n>We introduce an adaptive messaging mechanism, using dynamic gating units, to deeply integrate node and edge features for more accurate extreme event forecasting.
  arXiv  Detail & Related papers  (2025-02-01T06:49:16Z)
• ReFine: Boosting Time Series Prediction of Extreme Events by Reweighting and Fine-tuning [0.0]
  Extreme events are of great importance since they represent impactive occurrences. accurately predicting these extreme events is challenging due to their rarity and irregularity. We propose two strategies, reweighting and fine-tuning, to tackle the challenge.
  arXiv  Detail & Related papers  (2024-09-21T19:29:29Z)
• ExtremeCast: Boosting Extreme Value Prediction for Global Weather Forecast [57.6987191099507]
  We introduce Exloss, a novel loss function that performs asymmetric optimization and highlights extreme values to obtain accurate extreme weather forecast. We also introduce ExBooster, which captures the uncertainty in prediction outcomes by employing multiple random samples. Our solution can achieve state-of-the-art performance in extreme weather prediction, while maintaining the overall forecast accuracy comparable to the top medium-range forecast models.
  arXiv  Detail & Related papers  (2024-02-02T10:34:13Z)
• Performative Time-Series Forecasting [64.03865043422597]
  We formalize performative time-series forecasting (PeTS) from a machine-learning perspective.<n>We propose a novel approach, Feature Performative-Shifting (FPS), which leverages the concept of delayed response to anticipate distribution shifts.<n>We conduct comprehensive experiments using multiple time-series models on COVID-19 and traffic forecasting tasks.
  arXiv  Detail & Related papers  (2023-10-09T18:34:29Z)
• SEEDS: Emulation of Weather Forecast Ensembles with Diffusion Models [13.331224394143117]
  Uncertainty quantification is crucial to decision-making. dominant approach to representing uncertainty in weather forecasting is to generate an ensemble of forecasts. We propose to amortize the computational cost by emulating these forecasts with deep generative diffusion models learned from historical data.
  arXiv  Detail & Related papers  (2023-06-24T22:00:06Z)
• AA-Forecast: Anomaly-Aware Forecast for Extreme Events [25.89754218631525]
  Time series models often deal with extreme events and anomalies, both prevalent in real-world datasets. We propose an anomaly-aware forecast framework that leverages the previously seen effects of anomalies to improve its prediction accuracy. Specifically, the framework automatically extracts anomalies and incorporates them through an attention mechanism to increase its accuracy for future extreme events.
  arXiv  Detail & Related papers  (2022-08-21T17:51:46Z)
• Improving Event Duration Prediction via Time-aware Pre-training [90.74988936678723]
  We introduce two effective models for duration prediction. One model predicts the range/unit where the duration value falls in (R-pred); and the other predicts the exact duration value E-pred. Our best model -- E-pred, substantially outperforms previous work, and captures duration information more accurately than R-pred.
  arXiv  Detail & Related papers  (2020-11-05T01:52:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.