2 papers
cs.CR2025
PrompTrend: Continuous Community-Driven Vulnerability Discovery and Assessment for Large Language Models
Tarek Gasmi, Ramzi Guesmi, Mootez Aloui +1
Static benchmarks fail to capture LLM vulnerabilities emerging through community experimentation in online forums. We present PrompTrend, a system that collects vulnerability data…
cs.CR2025
Bridging AI and Software Security: A Comparative Vulnerability Assessment of LLM Agent Deployment Paradigms
Tarek Gasmi, Ramzi Guesmi, Ines Belhadj +1
Large Language Model (LLM) agents face security vulnerabilities spanning AI-specific and traditional software domains, yet current research addresses these separately. This study b…