SOURCE-LINKED INTELLIGENCE

Defence against the dark arts of distillation: Unsupervised discovery of subliminal learning

arXiv · AI, language, vision and robotics · article · Oct 9, 2026 · UTC

When a language model is distilled from a teacher that shares its base parameters, the student can inherit behavioural traits of the teacher---including misalignment---even from data that is semantically unrelated to the trait (Cloud et al., 2026). All existing defences require knowing the trait. We remove this requirement and prove that a subliminally transferred trait can be recovered from the parameter updates alone, without supervision, by exploiting a single asymmetry: the trait is expressed consistently across diverse datasets generated by the teacher, whereas task-specific and shared ca

Read original source ↗ Open in workspace

recordType

paper

region

Global