E4GEN: Event-level Explainable Extreme-Enhanced Time-series Generation

2026-05-31

AI Summary

Existing time series generation methods struggle to accurately model extreme events, often resulting in distorted simulations of critical anomalous behaviors. To address this limitation, this work proposes E4GEN—an interpretable diffusion-based generative framework that enables precise and controllable synthesis of extreme events through three core components: an E-Activator that adaptively triggers control signals, an E-Predictor that generates sample-specific control signals via self-driven semantic prediction, and an E-Control module that hierarchically injects these semantic signals into the denoising process. E4GEN is the first method to achieve event-level interpretable, unsupervised extreme event generation, introducing novel mechanisms including self-driven semantic prediction, data-conditioned training, and noise-initialization sampling. Evaluated across six datasets and seventeen metrics, it consistently outperforms state-of-the-art approaches, demonstrating significant improvements in overall fidelity, extreme event reconstruction accuracy, and downstream task utility.

Abstract

Generating realistic time series is essential for scientific research and real-world applications. However, existing methods often emphasize overall distributional fidelity while failing to faithfully capture extreme events. To advance existing research, we propose E4GEN, an explainable diffusion framework for extreme event-aware time-series generation. E4GEN provides systematic insights into when, what, and how to control extreme-event generation through three key components. First, E-Activator learns the dataset-adaptive extreme-control signal activation step during the denoising process without interfering with regular temporal components, including trend and seasonality. Second, E-Predictor determines what control signal to enforce through Self-Driven Semantic Prediction, where each sample derives its own control signal by inferring latent extreme-event information during generation. It also includes a novel Data-Conditioned Training, Noise-Initiated Sampling mechanism to address the issue of unavailable training labels. Third, E-Control specifies how to control extreme-event generation through a trainable Extreme Control Network, which transforms the semantic control signal into layer-wise signals and injects it into the denoising process. We evaluate E4GEN on six datasets with 17 metrics, and extensive experiments show that E4GEN outperforms state-of-the-art models across multiple dimensions, including overall fidelity, extreme-event fidelity, and downstream utility.

Problem

• time-series generation
• extreme events
• distributional fidelity
• event-level modeling
• realistic synthesis

Innovation

• extreme-event generation
• diffusion model
• explainable time-series generation
• self-driven semantic prediction
• data-conditioned training