# Debiased-CAM to mitigate image perturbations with faithful visual explanations of machine learning

arXiv: https://arxiv.org/abs/2012.05567 · HTML: https://mlsanity.com/paper/2012.05567/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 10 Dec 2020.

## Built on by (most cited first, 11 of 11)

- [Towards Human-centered Explainable AI: A Survey of User Studies for Model Explanations](https://mlsanity.com/paper/2210.11584/) (2022)
- [A Systematic Literature Review of Human-Centered, Ethical, and Responsible AI](https://mlsanity.com/paper/2302.05284/) (2023)
- [How explainable AI affects human performance: A systematic review of the behavioural consequences of saliency maps](https://mlsanity.com/paper/2404.16042/) (2024)
- [Statistically Significant Concept-based Explanation of Image Classifiers via Model Knockoffs](https://mlsanity.com/paper/2305.18362/) (2023)
- [On the Design and Evaluation of Human-centered Explainable AI Systems: A Systematic Review and Taxonomy](https://mlsanity.com/paper/2510.12201/) (2025)
- [Comparables XAI: Faithful Example-based AI Explanations with Counterfactual Trace Adjustments](https://mlsanity.com/paper/2602.13784/) (2026)
- [Beyond XAI:Obstacles Towards Responsible AI](https://mlsanity.com/paper/2309.03638/) (2023)
- [TA3: Testing Against Adversarial Attacks on Machine Learning Models](https://mlsanity.com/paper/2410.05334/) (2024)
- [SketchXplain: Intuitive Visual Explanations of Image Classifiers with Sketches](https://mlsanity.com/paper/2606.17646/) (2026)
- [Class Activation Mapping in Explainable Computer Vision: A Method-Centered Review of CNN, Transformer, and Foundation-Model-Era Visual Explanations](https://mlsanity.com/paper/2608.12299/) (2026)
- [AlignFace: Human-Aligned Face Similarity Metric with Interpretable Concept Relations](https://mlsanity.com/paper/2608.14130/) (2026)

## Abstract (arXiv)

Model explanations such as saliency maps can improve user trust in AI by highlighting important features for a prediction. However, these become distorted and misleading when explaining predictions of images that are subject to systematic error (bias) by perturbations and corruptions. Furthermore, the distortions persist despite model fine-tuning on images biased by different factors (blur, color temperature, day/night). We present Debiased-CAM to recover explanation faithfulness across various bias types and levels by training a multi-input, multi-task model with auxiliary tasks for explanation and bias level predictions. In simulation studies, the approach not only enhanced prediction accuracy, but also generated highly faithful explanations about these predictions as if the images were unbiased. In user studies, debiased explanations improved user task performance, perceived truthfulness and perceived helpfulness. Debiased training can provide a versatile platform for robust performance and explanation faithfulness for a wide range of applications with data biases.