# FAR: A General Framework for Attributional Robustness

arXiv: https://arxiv.org/abs/2010.07393 · HTML: https://mlsanity.com/paper/2010.07393/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Oct 2020.

## Built on by (most cited first, 13 of 13)

- [Robust Explanation Constraints for Neural Networks](https://mlsanity.com/paper/2212.08507/) (2022)
- [Manifold Integrated Gradients: Riemannian Geometry for Feature Attribution](https://mlsanity.com/paper/2405.09800/) (2024)
- [Fooling Explanations in Text Classifiers](https://mlsanity.com/paper/2206.03178/) (2022)
- [Exploiting the Relationship Between Kendall's Rank Correlation and Cosine Similarity for Attribution Protection](https://mlsanity.com/paper/2205.07279/) (2022)
- [On the Robustness of Explanations of Deep Neural Network Models: A Survey](https://mlsanity.com/paper/2211.04780/) (2022)
- [Towards More Robust Interpretation via Local Gradient Alignment](https://mlsanity.com/paper/2211.15900/) (2022)
- [A Practical Upper Bound for the Worst-Case Attribution Deviations](https://mlsanity.com/paper/2303.00340/) (2023)
- [Estimating the Adversarial Robustness of Attributions in Text with Transformers](https://mlsanity.com/paper/2212.09155/) (2022)
- [Provable Robust Saliency-based Explanations](https://mlsanity.com/paper/2212.14106/) (2022)
- [DARE: Towards Robust Text Explanations in Biomedical and Healthcare Applications](https://mlsanity.com/paper/2307.02094/) (2023)
- [Certified ℓ2 Attribution Robustness via Uniformly Smoothed Attributions](https://mlsanity.com/paper/2405.06361/) (2024)
- [Robust Ranking Explanations](https://mlsanity.com/paper/2307.04024/) (2023)
- [Improving Adversarial Robustness of Attribution via Implicit Regularization](https://mlsanity.com/paper/2605.29983/) (2026)

## Abstract (arXiv)

Attribution maps are popular tools for explaining neural networks predictions. By assigning an importance value to each input dimension that represents its impact towards the outcome, they give an intuitive explanation of the decision process. However, recent work has discovered vulnerability of these maps to imperceptible adversarial changes, which can prove critical in safety-relevant domains such as healthcare. Therefore, we define a novel generic framework for attributional robustness (FAR) as general problem formulation for training models with robust attributions. This framework consist of a generic regularization term and training objective that minimize the maximal dissimilarity of attribution maps in a local neighbourhood of the input. We show that FAR is a generalized, less constrained formulation of currently existing training methods. We then propose two new instantiations of this framework, AAT and AdvAAT, that directly optimize for both robust attributions and predictions. Experiments performed on widely used vision datasets show that our methods perform better or comparably to current ones in terms of attributional robustness while being more generally applicable. We finally show that our methods mitigate undesired dependencies between attributional robustness and some training and estimation parameters, which seem to critically affect other competitor methods.