Benchmarking Large Language Models for Log Analysis, Security, and Interpretation

Egil Karlsen; Xiao Luo; Nur Zincir-Heywood; Malcolm Heywood

Home/Archive/arXiv

arXiv · 2311.14519

Benchmarking Large Language Models for Log Analysis, Security, and Interpretation

Egil Karlsen·Xiao Luo·Nur Zincir-Heywood·Malcolm Heywood

View original source ↗Read source document ↗Export metadata ↓

Abstract

Large Language Models (LLM) continue to demonstrate their utility in a variety of emergent capabilities in different fields. An area that could benefit from effective language understanding in cybersecurity is the analysis of log files. This work explores LLMs with different architectures (BERT, RoBERTa, DistilRoBERTa, GPT-2, and GPT-Neo) that are benchmarked for their capacity to better analyze application and system log files for security. Specifically, 60 fine-tuned language models for log analysis are deployed and benchmarked. The resulting models demonstrate that they can be used to perform log analysis effectively with fine-tuning being particularly important for appropriate domain adaptation to specific log types. The best-performing fine-tuned sequence classification model (DistilRoBERTa) outperforms the current state-of-the-art; with an average F1-Score of 0.998 across six datasets from both web application and system log sources. To achieve this, we propose and implement a new experimentation pipeline (LLM4Sec) which leverages LLMs for log analysis experimentation, evaluation, and analysis.

Explore related subjects

cs.NI

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Egil Karlsen, Xiao Luo, Nur Zincir-Heywood, Malcolm Heywood. 2023-11-24. Benchmarking Large Language Models for Log Analysis, Security, and Interpretation. https://arxiv.org/abs/2311.14519

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

CSI Simulation: Why Additive Noise Fails and How to Fix It

Channel State Information (CSI) has become a widely used wireless channel sensing modality for applications such as indoor localization, activity recognition, and respiration monitoring. Because collecting labeled data under every target condition is impractical, training CSI-based models often relies on simulated data produced by adding noise or perturbations to recorded channel estimates, most commonly additive white Gaussian noise (AWGN). This practice assumes that the receiver chain between the antenna and the channel estimator is linear and gain-invariant. We test this assumption empirically using RF jamming as a controlled perturbation on 6 commodity receivers across 2 indoor environments. The assumption does not hold. Automatic gain control compresses the channel estimate multiplicatively before digitization, producing amplitude distributions that no additive noise variance can reproduce. To close the resulting fidelity gap, we propose M_QTC, a measurement-calibrated model that learns the per-subcarrier distribution transformation through quantile mapping, temporal filtering, and copula-based cross-subcarrier reordering. M_QTC reduces amplitude error 8-fold and closes 89% of the aggregate fidelity gap across four complementary dimensions. The improvement transfers directly to downstream tasks, where 5 classifiers from different families trained on M_QTC-simulated data recover 93% of real-data jamming detection performance, while AWGN-trained classifiers remain near random decision.

cs.NI↗

arXiv2026

BALANCE: Hybrid Autoregressive-Speculative LLM Inference at the Network Edge

Edge inference is a promising paradigm to provide large language model (LLM) inference services in next-generation mobile networks. LLM inference mainly relies on two approaches: Autoregressive decoding (AD) generates output tokens sequentially, resulting in long latency; Speculative decoding (SD) accelerates inference by using a small language model (SLM) to generate multiple draft tokens for LLM verification, but incurs extra memory costs. Due to this latency-memory tradeoff, neither approach alone can efficiently serve users with heterogeneous demands under limited edge computing resources. To address this challenge, we propose a hybrid autoregressive-speculative inference (BALANCE) framework for edge LLM inference. In BALANCE, an edge server hosts both an SLM and an LLM, admits users, assigns each admitted user to the AD or SD mode, and performs the two modes simultaneously. To maximize the number of served users, we formulate a task throughput maximization problem to jointly determine user admission and computing resource allocation between AD and SD under user latency requirements and server memory constraints. Since the problem is NP-hard, we develop a polynomial-time algorithm that transforms the original problem into two sub-problems and obtains a sub-optimal solution with a constant approximation guarantee. Experiments demonstrate that BALANCE consistently outperforms conventional AD and SD and significantly improves task throughput.

cs.NI↗

arXiv2026

Zero-Knowledge Remote Adversarial Attack against Wi-Fi-based Human Activity Recognition for Privacy Protection

The growing capability of Wi-Fi devices to identify human activities using channel state information (CSI) raises privacy concerns. To counter this threat, we propose GRAW, an adversary system, acting as a privacy defender, that degrades the human activity recognition (HAR) system at the user device by perturbing the router's signals that the device uses to estimate CSI. GRAW employs generative adversarial imitation learning (GAIL) to construct perturbation signals, and thereby eliminates the need for any information on the target HAR systems and their inputs (i.e., zero-knowledge operation). We evaluate GRAW against seven representative HAR models, using datasets collected in five environments, including our own dataset. We observe that GRAW is the only remote attack scheme that degrades every tested HAR model to a random-selection level. At the same perturbation level, GRAW achieves an attack success ratio up to 76.7% higher than comparison methods, while maintaining over 99% packet success rate on regular Wi-Fi communication. We demonstrate the feasibility of GRAW through real-time, over-the-air experiments with software-defined radios.

cs.NI↗