AI RESEARCH
KoALA: KL-L0 Adversarial Detector via Label Agreement
arXiv CS.LG
•
ArXi:2510.12752v2 Announce Type: replace Deep neural networks are highly susceptible to adversarial attacks, which pose significant risks to security- and safety-critical applications. We present KoALA (KL-L0 Adversarial detection via Label Agreement), a novel, semantics-free adversarial detector that requires no architectural changes or adversarial re