English(EN) FAME: Failure-Aware Mixture-of-Experts for Message-Level Log Anomaly Detection

FAME框架利用LLM实现高效日志异常检测

作者 PulseAugur 编辑部 · [2 个来源] · 2026-05-21 17:34

研究人员开发了FAME，一种用于消息级日志异常检测的新型框架，显著减少了手动标记的需求。该系统采用专家混合方法，利用大型语言模型离线将日志模板划分为故障域。FAME训练轻量级路由器和专家模型，这些模型可以在本地运行，在BGL和Thunderbird等基准数据集上取得高F1分数，同时大幅减少标注工作量。 AI

影响通过减少对大量手动标记的依赖，实现生产系统中更高效、更精确的异常检测。

排序理由该集群描述了一篇详细介绍日志异常检测新框架的新研究论文。

在 arXiv cs.LG 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。我们如何撰写摘要 →

报道来源 [2]

arXiv cs.LG TIER_1 English(EN) · Huanchi Wang, Zihang Huang, Yifang Tian, Kristina Dzeparoska, Hans-Arno Jacobsen, Alberto Leon-Garcia · 2026-05-22 04:00

FAME: Failure-Aware Mixture-of-Experts for Message-Level Log Anomaly Detection

arXiv:2605.22779v1 Announce Type: cross Abstract: Production systems generate millions of log lines daily, yet most anomaly detectors operate at the session or window-level, flagging groups of lines rather than identifying the specific message responsible. This coarse granularity…
arXiv cs.LG TIER_1 English(EN) · Alberto Leon-Garcia · 2026-05-21 17:34

FAME: Failure-Aware Mixture-of-Experts for Message-Level Log Anomaly Detection

Production systems generate millions of log lines daily, yet most anomaly detectors operate at the session or window-level, flagging groups of lines rather than identifying the specific message responsible. This coarse granularity forces operators to inspect many routine lines pe…

报道来源 [2]

FAME: Failure-Aware Mixture-of-Experts for Message-Level Log Anomaly Detection

FAME: Failure-Aware Mixture-of-Experts for Message-Level Log Anomaly Detection

相关实体

相关话题