LATIDIA · Ciberseguridad

Hacia la comprensión de la detección de anomalías de registro basadas en LLM: un estudio empírico de rendimiento, eficiencia y robustez

arXiv: 2609.31371v1Tipo de anuncio: Cross Resumen: Los modelos de lenguaje grandes (LLM) han demostrado un rendimiento prometedor en la detección de anomalías de registro, pero cómo se configuran sus estrategias de adaptación, arquitecturas y despliegue

La noticia

arXiv:2609.31371v1 Announce Type: cross Abstract: Large language models (LLMs) have demonstrated promising performance in log anomaly detection, yet how their adaptation strategies, architectures, and deployment configurations affect detection effectiveness remains insufficiently understood. To investigate these factors, we conduct a systematic empirical analysis across three public log datasets, examining different adaptation strategies, model architectures, parameter scales, and quantization settings. Our results reveal substantial performance differences across adaptation strategies, while model scaling yields varying detection gains across datasets. We further observe that models with comparable detection accuracy can exhibit markedly different computational costs, and that low-bit quantization largely preserves detection performance in the evaluated configurations. Finally, we examine detection robustness under structural,