Robustness of Llama-3.1-8B in Zero-Shot CWE Detection with Extended Context Lengths on Adversarial Python Code

Large Language Models (LLMs) have demonstrated significant capabilities in understanding and analyzing code for security vulnerabilities, such as Common Weakness Enumerations (CWEs). However, their reliance on cloud infrastructure and substantial computational requirements pose challenges for analyzing sensitive or proprietary codebases due to privacy concerns and inference costs. This work explores the potential of Small Language Models (SLMs) as a viable alternative for accurate, on-premise vulnerability detection. We investigated whether a 350-million parameter pre-trained code model (codeg

Research goal: What is the impact of extended context length on the robustness of Llama-3.1-8B's zero-shot CWE detection when tested on adversarially perturbed Python code from the Big-Vul dataset, measured by the drop in accuracy compared to unmodified code?

Autonomous synthesis report generated by SOVEREIGN Research Kernel. Tribunal consensus score: 9.3/10.