analiza aiqbench
ASTER introduces a novel unsupervised framework for time-series anomaly detection that leverages latent space pseudo-anomaly generation and pre-trained Large Language Models to enhance temporal representations. By eliminating the need for handcrafted injections or fixed distance metrics, this approach addresses data scarcity issues common in industrial monitoring and cybersecurity applications using Transformer-based classification. The method establishes a new benchmark for LLM-integrated time-series analysis without relying on traditional reconstruction techniques that often fail with complex datasets.
ASTER: Latent Pseudo-Anomaly Generation for Unsupervised Time-Series Anomaly Detection
arXiv:2604.13924v3 Announce Type: replace-cross Abstract: Time-series anomaly detection (TSAD) is critical in domains such as industrial monitoring, healthcare, and cybersecurity, but it remains challenging due to rare and heterogeneous anomalies and the scarcity of labelled data. This scarcity makes unsupervised approaches predominant, yet existing methods often rely on reconstruction or forecasting, which struggle with complex data, or on embedding-based approaches that require domain-specific anomaly synthesis and fixed distance metrics. We propose ASTER, a framework that generates pseudo-anomalies directly in the latent space, avoiding handcrafted anomaly injections and the need for domain expertise. A latent-space decoder produces tailored pseudo-anomalies to train a Transformer-based anomaly classifier, while a pre-trained LLM enriches the temporal and contextual representations of this space. Experiments on three benchmark datasets show that ASTER achieves state-of-the-art performance and sets a new standard for LLM-based TSAD.
Abstract:Time-series anomaly detection (TSAD) is critical in domains such as industrial monitoring, healthcare, and cybersecurity, but it remains challenging due to rare and heterogeneous anomalies and the scarcity of labelled data. This scarcity makes unsupervised approaches predominant, yet existing methods often rely on reconstruction or forecasting, which struggle with complex data, or on embedding-based approaches that require domain-specific anomaly synthesis and fixed distance metrics. We propose ASTER, a framework that generates pseudo-anomalies directly in the latent space, avoiding handcrafted anomaly injections and the need for domain expertise. A latent-space decoder produces tailored pseudo-anomalies to train a Transformer-based anomaly classifier, while a pre-trained LLM enriches the temporal and contextual representations of this space. Experiments on three benchmark datasets show that ASTER achieves state-of-the-art performance and sets a new standard for LLM-based TSAD.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
arXiv:2604.13924 [cs.LG]
(or arXiv:2604.13924v3 [cs.LG] for this version)
Text integral sub licența deschisă a publicației · 151 cuvinte