Smart-LLaMA-DPO: Reinforced Large Language Model for Explainable Smart Contract Vulnerability Detection

Smart-LLaMA-DPO is an advanced smart contract vulnerability detection and explanation system built upon the LLaMA-3.1-8B model. The system addresses key challenges in smart contract security using a comprehensive approach that combines Continual Pre-training (CPT), Supervised Fine-tuning (SFT), and Direct Preference Optimization (DPO). This ensures enhanced performance in identifying vulnerabilities and generating high-quality explanations for detected issues.