Llama-based source code vulnerability detection: Prompt engineering vs Fine tuning

This article from ArXiv likely compares two methods for using Llama models to detect vulnerabilities in source code: prompt engineering and fine-tuning. The analysis would likely involve comparing the performance, efficiency, and potential drawbacks of each approach. The 'vs' in the title suggests a direct comparison and evaluation of the two techniques.