Workshop Program

Time	Program
09:00–09:10	Opening Remarks
Session 1: “The Opposite of Every Instinct I Have” Evaluation and Behavior Analysis of LLMs
09:10–09:25	A Temporal Paradox in Software Vulnerability Prioritization: Why Do Large Language Models Perform Better Post-Knowledge Cutoff Date? Osama Al Haddad, Muhammad Ikram, Young Choon Lee, Muhammad Ejaz Ahmed
09:25–09:40	On the Difficulty of Selecting Few-Shot Examples for Effective LLM-based Vulnerability Detection Md Abdul Hannan, Ronghao Ni, Chi Zhang, Limin Jia, Ravi Mangal, Corina S. Pasareanu
09:40–09:55	Local LLMs for NL2Bash: A Large-Scale Open-Source Model Evaluation for Bash Command Jef Jacobs, Jorn Lapon, Vincent Naessens
9:55–10:20	Morning Break
Session 2: “Because He’s My Butler!” LLM Agents and Assistants
10:20–10:35	AWE: Adaptive Agents for Dynamic Web Penetration Testing Akshat Singh Jaswal, Ashish Baghel
10:35–10:50	Context Relay for Long-Running Penetration-Testing Agents Marius Vangeli, Joel Brynielsson, Mika Cohen, Farzad Kamrani
10:50–11:05	Vibenix: An AI Assistant for Software Packaging with Nix Martin Schwaighofer, Martim Monis, Nuno Saavedra, João F. Ferreira, Rene Mayrhofer
Session 3: “And So We Danced” LLM-Based Attack and Defense
11:05–11:20	Proactive Hardening of LLM Defenses with HASTE Henry Chen, Victor Aranda, Samarth Keshari, Ryan Heartfield, Nicole Nichols
11:20–11:35	Auditable LLM Arbiter for DeFi Security: A Hybrid Graph-of-Thoughts Approach to Intent–Transaction Alignment Duanyi Yao, Siddhartha Jagannath, Baltasar Aroso, Vyas Krishnan, Ding Zhao
11:35–11:50	PriMod4AI: Lifecycle-Aware Privacy Threat Modeling for AI Systems using LLM Gautam Savaliya, Robert Aufschläger, Abhishek Subedi, Michael Heigl, Martin Schramm
11:50–12:05	HELIOS: Hierarchical Graph Abstraction for Structure-Aware LLM Decompilation Yonatan Gizachew Achamyeleh, Harsh Thomare, Mohammad Abdullah Al Faruque
12:05–13:30	Lunch Break
13:30–17:30	LLM for Security Competition