Frequency-Aware Model Parameter Explorer: A new attribution method for improving explainability

ArXi:2510.03245v2 Announce Type: replace-cross State-of-the-art attribution methods rely on adversarial sample generation that applies an all-pass filter across the frequency spectrum, discarding fine-grained high-frequency information that is nstrably important for accurate feature attribution in deep neural networks. By generating adversarial samples that selectively perturb high- and low-frequency components, we can probe which spectral features a model relies on most -- directly translating frequency-domain exploration into attribution signals.