# There is More to Attention: Statistical Filtering Enhances Explanations in Vision Transformers

arXiv: https://arxiv.org/abs/2510.06070 · HTML: https://mlsanity.com/paper/2510.06070/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 34 references that are in the map.
- First posted on arXiv 7 Oct 2025.

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

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networks](https://mlsanity.com/paper/1710.11063/) (2017)
- [Attention is not Explanation](https://mlsanity.com/paper/1902.10186/) (2019)
- [RISE: Randomized Input Sampling for Explanation of Black-box Models](https://mlsanity.com/paper/1806.07421/) (2018)
- [Quantifying Attention Flow in Transformers](https://mlsanity.com/paper/2005.00928/) (2020)
- [Do Vision Transformers See Like Convolutional Neural Networks?](https://mlsanity.com/paper/2108.08810/) (2021)
- [Attention is not not Explanation](https://mlsanity.com/paper/1908.04626/) (2019)
- [Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LM](https://mlsanity.com/paper/2104.04473/) (2021)
- [Transformer Interpretability Beyond Attention Visualization](https://mlsanity.com/paper/2012.09838/) (2020)
- [Towards Robust Interpretability with Self-Explaining Neural Networks](https://mlsanity.com/paper/1806.07538/) (2018)
- [Evaluating the visualization of what a Deep Neural Network has learned](https://mlsanity.com/paper/1509.06321/) (2015)
- [On the (In)fidelity and Sensitivity for Explanations](https://mlsanity.com/paper/1901.09392/) (2019)
- [Analyzing the Structure of Attention in a Transformer Language Model](https://mlsanity.com/paper/1906.04284/) (2019)
- [Layer-wise Relevance Propagation for Neural Networks with Local Renormalization Layers](https://mlsanity.com/paper/1604.00825/) (2016)
- [AttnLRP: Attention-Aware Layer-Wise Relevance Propagation for Transformers](https://mlsanity.com/paper/2402.05602/) (2024)
- [XAI for Transformers: Better Explanations through Conservative Propagation](https://mlsanity.com/paper/2202.07304/) (2022)
- [Analyzing the Source and Target Contributions to Predictions in Neural Machine Translation](https://mlsanity.com/paper/2010.10907/) (2020)
- [Deep Visual Attention Prediction](https://mlsanity.com/paper/1705.02544/) (2017)
- [Explainable AI for Safe and Trustworthy Autonomous Driving: A Systematic Review](https://mlsanity.com/paper/2402.10086/) (2024)
- [Decoupling Pixel Flipping and Occlusion Strategy for Consistent XAI Benchmarks](https://mlsanity.com/paper/2401.06654/) (2024)
- [Opti-CAM: Optimizing saliency maps for interpretability](https://mlsanity.com/paper/2301.07002/) (2023)
- [Attention Meets Post-hoc Interpretability: A Mathematical Perspective](https://mlsanity.com/paper/2402.03485/) (2024)
- [An Attention Matrix for Every Decision: Faithfulness-based Arbitration Among Multiple Attention-Based Interpretations of Transformers in Text Classification](https://mlsanity.com/paper/2209.10876/) (2022)
- [On the stability, correctness and plausibility of visual explanation methods based on feature importance](https://mlsanity.com/paper/2311.12860/) (2023)

## Built on by (most cited first, 1 of 1)

- [Interpreto: An Explainability Library for Transformers](https://mlsanity.com/paper/2512.09730/) (2025)

## Abstract (arXiv)

Explainable AI (XAI) has become increasingly important with the rise of large transformer models, yet many explanation methods designed for CNNs transfer poorly to Vision Transformers (ViTs). Existing ViT explanations often rely on attention weights, which tend to yield noisy maps as they capture token-to-token interactions within each layer.While attribution methods incorporating MLP blocks have been proposed, we argue that attention remains a valuable and interpretable signal when properly filtered. We propose a method that combines attention maps with a statistical filtering, initially proposed for CNNs, to remove noisy or uninformative patterns and produce more faithful explanations. We further extend our approach with a class-specific variant that yields discriminative explanations. Evaluation against popular state-of-the-art methods demonstrates that our approach produces sharper and more interpretable maps. In addition to perturbation-based faithfulness metrics, we incorporate human gaze data to assess alignment with human perception, arguing that human interpretability remains essential for XAI. Across multiple datasets, our approach consistently outperforms or is comparable to the SOTA methods while remaining efficient and human plausible.