All you need to break LLMs are Black-Box, Adapting, Efficient, Transferable, Harmful, Applicable... Attacks
Indirect Harm Optimization (IHO) is a novel black-box attacker framework designed to reliably evaluate LLM jailbreaks by optimizing for six criteria: efficiency, knowledge access, harmfulness, adaptiveness, transferability, and applicability. It uses iterative preference optimization against a harm judge to create an efficient attacker model that generalizes across different target models and behaviors.