arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2503.08045cs.LGcs.CL
收藏

适配大语言模型以实现参数高效的日志异常检测

Adapting Large Language Models for Parameter-Efficient Log Anomaly Detection

Singapore Management University(新加坡管理大学)

Ying Fu Lim, Jiawen Zhu, Guansong Pang

AI总结:

本文探索利用LoRA和ReFT等参数高效微调技术适配RoBERTa、GPT-2和Llama-3等大语言模型进行日志异常检测,并通过四个公开数据集的实验验证了方法的有效性、稳定性、样本效率、鲁棒性和跨数据集泛化能力。

摘要:

日志异常检测(LAD)旨在识别日志数据中的非典型模式,这些模式对于评估系统的安全性和状态至关重要。尽管大语言模型(LLMs)在众多领域取得了巨大成功,但利用LLMs实现日志异常检测在很大程度上仍未得到探索。本研究旨在填补这一空白。由于对LLMs进行完全微调涉及高昂的成本,我们探索使用参数高效微调技术(PEFTs)将LLMs适配到LAD任务。为了深入探索LLM驱动的LAD的潜力,我们对利用两种最流行的PEFTs——低秩适配(LoRA)和表示微调(ReFT)——进行了全面研究,以挖掘三种不同规模的知名LLMs,包括RoBERTa、GPT-2和Llama-3,用于参数高效的LAD。我们在四个公开日志数据集上进行了全面实验,从几个关键视角揭示了有效LLM驱动的LAD的重要见解,包括这些基于PEFT的LLM驱动LAD方法的有效性、稳定性、样本效率、对不稳定日志的鲁棒性以及跨数据集泛化能力。代码可在https://github.com/mala-lab/LogADReft获取。

英文摘要:

Log Anomaly Detection (LAD) seeks to identify atypical patterns in log data that are crucial to assessing the security and condition of systems. Although Large Language Models (LLMs) have shown tremendous success in various fields, the use of LLMs in enabling the detection of log anomalies is largely unexplored. This work aims to fill this gap. Due to the prohibitive costs involved in fully fine-tuning LLMs, we explore the use of parameter-efficient fine-tuning techniques (PEFTs) for adapting LLMs to LAD. To have an in-depth exploration of the potential of LLM-driven LAD, we present a comprehensive investigation of leveraging two of the most popular PEFTs -- Low-Rank Adaptation (LoRA) and Representation Fine-tuning (ReFT) -- to tap into three prominent LLMs of varying size, including RoBERTa, GPT-2, and Llama-3, for parameter-efficient LAD. Comprehensive experiments on four public log datasets are performed to reveal important insights into effective LLM-driven LAD in several key perspectives, including the efficacy of these PEFT-based LLM-driven LAD methods, their stability, sample efficiency, robustness w.r.t. unstable logs, and cross-dataset generalization. Code is available at https://github.com/mala-lab/LogADReft.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑