Helmut Neukirchen

.- AI Vulnerabilities..- How Real is Your Jailbreak? Fine-grained Jailbreak Evaluation with Anchored Reference..- MEEA: Mere Exposure Effect-Driven Confrontational Optimization for LLM Jailbreaking..- Exposing Vulnerabilities in AI-Based Malware Detectors: Adversarial, Backdoor, and Poisoning Attacks..- Detecting and Analysing GenAI Output..- FreDA: Training-Free Test-Time Adaptation for Deepfake Detection via Non-Parametric Cache Retrieval..- A Multi-stage Framework for Detection and Attribution of LLM-generated Text..- MARS: LLM Toxicity Mitigation via Multi-Dimensional Adaptive Representation Steering..- AI-Enhanced...