Phishsense-1B: A Technical Perspective on an AI-Powered Phishing Detection Model
This work addresses phishing threats for cybersecurity applications, presenting an incremental improvement over existing models.
The paper tackled phishing detection by adapting the Llama-Guard-3-1B model using LoRA and GuardReasoner finetuning, achieving 97.5% accuracy on a custom dataset and 70% accuracy on a real-world dataset, outperforming unadapted models and BERT-based detectors.
Phishing is a persistent cybersecurity threat in today's digital landscape. This paper introduces Phishsense-1B, a refined version of the Llama-Guard-3-1B model, specifically tailored for phishing detection and reasoning. This adaptation utilizes Low-Rank Adaptation (LoRA) and the GuardReasoner finetuning methodology. We outline our LoRA-based fine-tuning process, describe the balanced dataset comprising phishing and benign emails, and highlight significant performance improvements over the original model. Our findings indicate that Phishsense-1B achieves an impressive 97.5% accuracy on a custom dataset and maintains strong performance with 70% accuracy on a challenging real-world dataset. This performance notably surpasses both unadapted models and BERT-based detectors. Additionally, we examine current state-of-the-art detection methods, compare prompt-engineering with fine-tuning strategies, and explore potential deployment scenarios.