ProDG: Prototypes for Data-Free Generative Post-Hoc Explainability

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Borycki, Piotr, Trędowicz, Magdalena, Tabor, Jacek, Struski, Łukasz, Spurek, Przemysław
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914583038394368
author Borycki, Piotr
Trędowicz, Magdalena
Tabor, Jacek
Struski, Łukasz
Spurek, Przemysław
author_facet Borycki, Piotr
Trędowicz, Magdalena
Tabor, Jacek
Struski, Łukasz
Spurek, Przemysław
contents Ante-hoc interpretability methods based on prototypes provide highly accurate explanations by utilizing the intuitive "this looks like that" reasoning paradigm. On the other hand, post-hoc models can explain predictions for a single image without relying on an underlying dataset or requiring costly neural network retraining. Recent approaches successfully solve the retraining problem for prototype-based networks. However, they still face a fundamental limitation: they require access to a subset of data (e.g., a test or validation set) to search for and extract the visual prototypes. In this paper, we address this issue and introduce ProDG: Generative Prototypes for Data-Free Post-Hoc Explainability, a novel framework that leverages generative models to synthesize pure, high-fidelity prototypes directly from the frozen model's weights, completely eliminating the dependency on any external data. By establishing this new frontier in Data-Free XAI, ProDG unlocks robust visual interpretability for privacy-sensitive domains, where original data is strictly restricted or fundamentally inaccessible. Project page: https://github.com/piotr310100/ProDG
format Preprint
id arxiv_https___arxiv_org_abs_2605_08858
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle ProDG: Prototypes for Data-Free Generative Post-Hoc Explainability
Borycki, Piotr
Trędowicz, Magdalena
Tabor, Jacek
Struski, Łukasz
Spurek, Przemysław
Computer Vision and Pattern Recognition
Ante-hoc interpretability methods based on prototypes provide highly accurate explanations by utilizing the intuitive "this looks like that" reasoning paradigm. On the other hand, post-hoc models can explain predictions for a single image without relying on an underlying dataset or requiring costly neural network retraining. Recent approaches successfully solve the retraining problem for prototype-based networks. However, they still face a fundamental limitation: they require access to a subset of data (e.g., a test or validation set) to search for and extract the visual prototypes. In this paper, we address this issue and introduce ProDG: Generative Prototypes for Data-Free Post-Hoc Explainability, a novel framework that leverages generative models to synthesize pure, high-fidelity prototypes directly from the frozen model's weights, completely eliminating the dependency on any external data. By establishing this new frontier in Data-Free XAI, ProDG unlocks robust visual interpretability for privacy-sensitive domains, where original data is strictly restricted or fundamentally inaccessible. Project page: https://github.com/piotr310100/ProDG
title ProDG: Prototypes for Data-Free Generative Post-Hoc Explainability
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2605.08858