# Diffusion Visual Counterfactual Explanations

Source: [Diffusion Visual Counterfactual Explanations](<https://www.sota2.com/research/paper/68fa6dbf1e209131cea3bd54>)

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

Data updated: 2025-12-24T17:57:56.764Z

Authors: Maximilian Augustin, Valentyn Boreiko, Francesco Croce, Matthias Hein

Published: 2022-10-21T09:35:47.000Z

## Abstract

Visual Counterfactual Explanations (VCEs) are an important tool to understand the decisions of an image classifier. They are 'small' but 'realistic' semantic changes of the image changing the classifier decision. Current approaches for the generation of VCEs are restricted to adversarially robust models and often contain non-realistic artefacts, or are limited to image classification problems with few classes. In this paper, we overcome this by generating Diffusion Visual Counterfactual Explanations (DVCEs) for arbitrary ImageNet classifiers via a diffusion process. Two modifications to the diffusion process are key for our DVCEs: first, an adaptive parameterization, whose hyperparameters generalize across images and models, together with distance regularization and late start of the diffusion process, allow us to generate images with minimal semantic changes to the original ones but different classification. Second, our cone regularization via an adversarially robust model ensures that the diffusion process does not converge to trivial non-semantic changes, but instead produces realistic images of the target class which achieve high confidence by the classifier.

## 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.

- [Counterfactual Explanation on ImageNet Zebra - Sorrel](<https://www.sota2.com/research/sota/counterfactual-explanation-on-imagenet-zebra-sorrel.md>): Counterfactual Explanation; ImageNet Zebra - Sorrel; FID: 33.1.

- [Counterfactual Explanation on ImageNet (Cheetah - Cougar)](<https://www.sota2.com/research/sota/counterfactual-explanation-on-imagenet-cheetah-cougar.md>): Counterfactual Explanation; ImageNet (Cheetah - Cougar); FID: 46.9.

- [Counterfactual Explanation on ImageNet Egyptian Cat - Persian Cat](<https://www.sota2.com/research/sota/counterfactual-explanation-on-imagenet-egyptian-cat-persian-cat.md>): Counterfactual Explanation; ImageNet Egyptian Cat - Persian Cat; FID: 46.6.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2210.11841>)

[PDF](<https://arxiv.org/pdf/2210.11841.pdf>)

[Implementation](<https://github.com/valentyn1boreiko/DVCEs>)