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

Source: [ProDG: Prototypes for Data-Free Generative Post-Hoc Explainability](<https://www.sota2.com/research/paper/6a02b406dbc836b87d824603>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-05-22T05:00:12.726Z

Authors: Piotr Borycki, Magdalena Tr\\k{e}dowicz, Jacek Tabor, {\\L}ukasz Struski, Przemys{\\l}aw Spurek

Published: 2026-05-09T10:14:46.000Z

## Abstract

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

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Explanation Matching on ImageNet User Study (test)](<https://www.sota2.com/research/sota/explanation-matching-on-imagenet-user-study-test.md>): Explanation Matching; ImageNet User Study (test); Accuracy: 73.

- [Prototype Discriminability on ImageNet](<https://www.sota2.com/research/sota/prototype-discriminability-on-imagenet.md>): Prototype Discriminability; ImageNet; Accuracy: 73.

- [Explanation Matching on CUB-200 User Study 2011 (test)](<https://www.sota2.com/research/sota/explanation-matching-on-cub-200-user-study-2011-test.md>): Explanation Matching; CUB-200 User Study 2011 (test); Accuracy: 37.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2605.08858>)

[PDF](<https://arxiv.org/pdf/2605.08858.pdf>)