FAME: Failure-Aware Mixture-of-Experts for Message-Level Log Anomaly Detection
FAME:面向失败的混合专家模型用于消息级日志异常检测
机构 * Department of Electrical and Computer Engineering(电气与计算机工程系)
专题命中 效率与部署 :LLM(abstract,abstract_cn);large language model(abstract);language model(abstract);分类 cs.LG
AI总结 本文提出FAME,一种面向失败的混合专家模型,用于消息级日志异常检测。该方法通过少量标注数据训练轻量级路由器和领域专家,实现高效的异常检测,同时在BGL和Thunderbird数据集上取得了高精度和召回率。
Comments 12 pages, 5 figures, Accepted at the 2026 IEEE International Symposium on Software Reliability Engineering (ISSRE 2026)