Search papers, labs, and topics across Lattice.
This study evaluates the fine-tuning of open-source large language models (LLMs) ranging from 2 to 9 billion parameters for cybersecurity and IT support, employing three techniques: Supervised Fine-Tuning (SFT), Low-Rank Adaptation (LoRA), and Quantised Low-Rank Adaptation (QLoRA). The findings reveal that while SFT excels in convergence metrics, both LoRA and QLoRA drastically reduce computational costs鈥攂y 99% and 60%, respectively鈥攚ithout sacrificing performance. Models like Llama, Mistral, and Phi demonstrate robust generalization across these methods, highlighting their potential for deployment in environments with limited resources.
SFT outperforms other fine-tuning methods in loss convergence while LoRA and QLoRA slash computational costs, making LLMs viable for resource-constrained cybersecurity applications.
The focus of Artificial Intelligence (AI) has progressed from exploring whether Large Language Models (LLMs) can transform domain-specific tasks to understanding how to optimise their impact. Advances in Natural Language Processing (NLP) have enabled LLMs to address complex tasks such as question answering, text classification, and translation with notable accuracy and efficiency. Hence, this study evaluates the effectiveness of fine-tuning open-source LLMs, which range from 2 to 9 billion parameters, for cybersecurity and IT support using three advanced techniques: Supervised Fine-Tuning (SFT), Low-Rank Adaptation (LoRA), and Quantised Low-Rank Adaptation (QLoRA). The results indicate that SFT achieves superior convergence in loss metrics for both tasks, while LoRA and QLoRA significantly reduce GPU memory and computational costs by 99% and 60%, respectively, with comparable performance. Models such as Llama, Mistral, and Phi consistently demonstrated high generalisation across all methods, making them particularly suitable for deployment in resource-constrained environments.
