When the harmfulness of an LLM agent's output can be quantified, a natural jailbreaking objective is to maximize expected harmfulness over admissible input modifications. An alternative approach constructs or selects harmful target outputs and modifies the input to increase their likelihood. We establish a precise conn …
Subliminal learning is a phenomenon where a student language model acquires a teacher model's behavioral traits by training on semantically unrelated outputs. It is a subtle statistical phenomenon as trait transmission relies on weak statistical patterns in the generated data. To understand trait transmission between t …
Strong unrestricted adversarial attacks can distort the primary object of an image, hereafter referred to as the subject. To preserve subject integrity without compromising attack magnitude, we introduce the carrier: a secondary visual element that provides an auxiliary region to facilitate the attack under global clas …
Building on a probabilistic perspective in which adversarial examples arise from the overlap between a distance-based distribution $p_{\mathrm{dis}}$ and a victim-classifier-induced distribution $p_{\mathrm{vic}}$, we start from a simple intuition: adversarial examples become harder to generate when these two distribut …