# Improving the Perturbation-Based Explanation of Deepfake Detectors Through the Use of Adversarially-Generated Samples

arXiv: https://arxiv.org/abs/2502.03957 · HTML: https://mlsanity.com/paper/2502.03957/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 40 references that are in the map.
- First posted on arXiv 6 Feb 2025.

## Builds on (most cited first, 25 of 25)

- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- ["Why Should I Trust You?": Explaining the Predictions of Any Classifier](https://mlsanity.com/paper/1602.04938/) (2016)
- [Xception: Deep Learning with Depthwise Separable Convolutions](https://mlsanity.com/paper/1610.02357/) (2016)
- [EfficientNetV2: Smaller Models and Faster Training](https://mlsanity.com/paper/2104.00298/) (2021)
- [FaceForensics++: Learning to Detect Manipulated Facial Images](https://mlsanity.com/paper/1901.08971/) (2019)
- [Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networks](https://mlsanity.com/paper/1710.11063/) (2017)
- [RISE: Randomized Input Sampling for Explanation of Black-box Models](https://mlsanity.com/paper/1806.07421/) (2018)
- [The DeepFake Detection Challenge (DFDC) Dataset](https://mlsanity.com/paper/2006.07397/) (2020)
- [A Review of Speaker Diarization: Recent Advances with Deep Learning](https://mlsanity.com/paper/2101.09624/) (2021)
- [A Consistent and Efficient Evaluation Strategy for Attribution Methods](https://mlsanity.com/paper/2202.00449/) (2022)
- [Video Summarization Using Deep Neural Networks: A Survey](https://mlsanity.com/paper/2101.06072/) (2021)
- [A Survey on Text Classification: From Shallow to Deep Learning](https://mlsanity.com/paper/2008.00364/) (2020)
- [Can ChatGPT Detect DeepFakes? A Study of Using Multimodal Large Language Models for Media Forensics](https://mlsanity.com/paper/2403.14077/) (2024)
- [Adversarial Deepfakes: Evaluating Vulnerability of Deepfake Detectors to Adversarial Examples](https://mlsanity.com/paper/2002.12749/) (2020)
- [Look at the Variance! Efficient Black-box Explanations with Sobol-based Sensitivity Analysis](https://mlsanity.com/paper/2111.04138/) (2021)
- [Towards Explainable Deep Neural Networks (xDNN)](https://mlsanity.com/paper/1912.02523/) (2019)
- [Foundation Models for Video Understanding: A Survey](https://mlsanity.com/paper/2405.03770/) (2024)
- [Explainable Artificial Intelligence: A Survey of Needs, Techniques, Applications, and Future Direction](https://mlsanity.com/paper/2409.00265/) (2024)
- [Metrics for saliency map evaluation of deep learning explanation methods](https://mlsanity.com/paper/2201.13291/) (2022)
- [An adversarial attack approach for eXplainable AI evaluation on deepfake detection models](https://mlsanity.com/paper/2312.06627/) (2023)
- [Towards Quantitative Evaluation of Explainable AI Methods for Deepfake Detection](https://mlsanity.com/paper/2404.18649/) (2024)
- [T-TAME: Trainable Attention Mechanism for Explaining Convolutional Networks and Vision Transformers](https://mlsanity.com/paper/2403.04523/) (2024)
- [What's wrong with this video? Comparing Explainers for Deepfake Detection](https://mlsanity.com/paper/2105.05902/) (2021)

## Built on by (most cited first, 2 of 2)

- [Unmasking Synthetic Realities in Generative AI: A Comprehensive Review of Adversarially Robust Deepfake Detection Systems](https://mlsanity.com/paper/2507.21157/) (2025)
- [An Experimental Study on Generating Plausible Textual Explanations for Video Summarization](https://mlsanity.com/paper/2509.26225/) (2025)

## Abstract (arXiv)

In this paper, we introduce the idea of using adversarially-generated samples of the input images that were classified as deepfakes by a detector, to form perturbation masks for inferring the importance of different input features and produce visual explanations. We generate these samples based on Natural Evolution Strategies, aiming to flip the original deepfake detector's decision and classify these samples as real. We apply this idea to four perturbation-based explanation methods (LIME, SHAP, SOBOL and RISE) and evaluate the performance of the resulting modified methods using a SOTA deepfake detection model, a benchmarking dataset (FaceForensics++) and a corresponding explanation evaluation framework. Our quantitative assessments document the mostly positive contribution of the proposed perturbation approach in the performance of explanation methods. Our qualitative analysis shows the capacity of the modified explanation methods to demarcate the manipulated image regions more accurately, and thus to provide more useful explanations.