Oral Session
in
Workshop: AdvML-Frontiers × CoTMA: From Model Security to Compositional Threats in Multi-Agent AI Systems Fri, Oct 9, 2026 • 11:45 AM – 12:00 PM PDT Franciscan C&D / Ballroom Level

Oral/Spotlight Paper Presentations I

Abstract

Auditing Prompt Injection Defenses Before They Reach Users: What Strong Adversaries Reveal Xiaoxue Yang, Bozhidar Stevanoski, Matthieu Meeus, Yves-Alexandre de Montjoye

Chat is not available.