Quantifying Prototype Stability in ProtoPNet Without Manual Part Annotations

Prototype-based networks such as ProtoPNet offer intrinsically inter-
pretable predictions by matching regions of the inferenced image to learned pro-
totypical patches. However, existing stability metrics rely on expensive manual
part annotations and are limited to narrow perturbation types. In this work, we
introduce the Structural Stability Score (Sss), a scalable, annotation-free metric
that quantifies prototype stability under a diverse set of visual transformations by
comparing prototype activation maps. We evaluate Sss on two ProtoPNet variants
(using VGG19 and Resnet34 backbones) trained on the CUB-200-2011 dataset,
and assess stability across six distinct perturbations. Our results reveal clear dif-
ferences in robustness both between models and among different transformations.
These findings demonstrate that Sss is a practical tool for highlighting stability
variations within and across prototype-based networks, guiding model selection
and interpretability analysis.