Lwin Khin Shar

The paper demonstrates that security patch detection models trained solely on publicly reported vulnerabilities (NVD) perform poorly when tested on real-world, unreported 'in-the-wild' patches, suggesting the need for diverse training data.

The paper introduces ASTRAL, a multimodal LLM-driven framework that reconstructs and analyzes fragmented cyber-physical system architectures to enable comprehensive and quantitative security risk assessment.

TitanCA presents a novel, multi-agent LLM orchestration framework that significantly improves vulnerability discovery by reducing false positives and identifying numerous zero-day vulnerabilities.