OpenSOC-AI: Democratizing Security Operations with Parameter Efficient LLM Log Analysis
Chaitanya Vilas Garware, Sharif Noor Zisad
TLDR
OpenSOC-AI uses parameter-efficient LLM fine-tuning to provide automated security log analysis for SMBs, democratizing threat detection.
Key contributions
- Introduces OpenSOC-AI, a lightweight framework for automated security log analysis.
- Employs parameter-efficient LoRA fine-tuning of TinyLlama-1.1B on a single T4 GPU.
- Performs automated threat classification, MITRE ATT&CK mapping, and severity assessment.
- Achieves significant accuracy gains (up to 68%) in threat detection over untuned baselines.
Why it matters
Small and medium businesses often lack resources for robust cybersecurity operations. OpenSOC-AI provides an accessible, efficient LLM-based solution for automated threat detection and analysis. This democratizes advanced security operations, making them viable for smaller organizations.
Original Abstract
Small and medium sized businesses (SMBs) face an escalating cybersecurity threat landscape, yet most lack the resources to staff full Security Operations Centers (SOCs) or deploy enterprise grade detection platforms. This paper presents OpenSOC-AI, a lightweight log analysis framework that uses parameter efficient fine tuning of a 1.1-billion parameter language model (TinyLlama-1.1B) to perform automated threat classification, MITRE ATT&CK technique mapping, and severity assessment on raw security log entries. Using Low-Rank Adaptation (LoRA) with only 12.6 million trainable parameters (roughly 1.13% of the base model), we fine tuned on 450 domain specific SOC examples in under five minutes on a single NVIDIA T4 GPU. Testing on a heldout set of 50 examples showed a 68% point gain in threat classification accuracy (from 0% to 68%), a 30% point gain in severity accuracy (from 28% to 58%), and an F1 score of 0.68 compared to the untuned baseline. Full codebase, adapter weights, and datasets are publicly released to support reproducibility and community extension.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.