WorksComputer Science
A Backdoor Trigger Is Legible in Fine-Tune Weights
Abstract
A backdoored language model behaves normally until a trigger phrase appears in its input. Every published method for recovering such a trigger runs the model. We show the trigger is legible in the weights alone: a static statistic reads a fine-tuned checkpoint with no forward pass of the suspect, no data, no labelled population of poisoned models and no prior knowledge that a backdoor exists, and names one of the tokens of the trigger phrase as the single highest-ranked candidate in the vocabulary.