# Beyond Output Faithfulness: Learning Attributions that Preserve Computational Pathways

arXiv: https://arxiv.org/abs/2509.04588 · HTML: https://mlsanity.com/paper/2509.04588/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 27 of its 30 references that are in the map.
- First posted on arXiv 4 Sep 2025.

## Builds on (most cited first, 27 of 27)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [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)
- [Axiomatic Attribution for Deep Networks](https://mlsanity.com/paper/1703.01365/) (2017)
- [Learning Deep Features for Discriminative Localization](https://mlsanity.com/paper/1512.04150/) (2015)
- [Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps](https://mlsanity.com/paper/1312.6034/) (2013)
- [Visualizing and Understanding Convolutional Networks](https://mlsanity.com/paper/1311.2901/) (2013)
- [Sparse Autoencoders Find Highly Interpretable Features in Language Models](https://mlsanity.com/paper/2309.08600/) (2023)
- [Sanity Checks for Saliency Maps](https://mlsanity.com/paper/1810.03292/) (2018)
- [SmoothGrad: removing noise by adding noise](https://mlsanity.com/paper/1706.03825/) (2017)
- [Striving for Simplicity: The All Convolutional Net](https://mlsanity.com/paper/1412.6806/) (2014)
- [Toy Models of Superposition](https://mlsanity.com/paper/2209.10652/) (2022)
- [Adversarial Examples Are Not Bugs, They Are Features](https://mlsanity.com/paper/1905.02175/) (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)
- [Network Dissection: Quantifying Interpretability of Deep Visual Representations](https://mlsanity.com/paper/1704.05796/) (2017)
- [Interpretable Explanations of Black Boxes by Meaningful Perturbation](https://mlsanity.com/paper/1704.03296/) (2017)
- [Score-CAM: Score-Weighted Visual Explanations for Convolutional Neural Networks](https://mlsanity.com/paper/1910.01279/) (2019)
- [Evaluating the visualization of what a Deep Neural Network has learned](https://mlsanity.com/paper/1509.06321/) (2015)
- [Understanding Deep Networks via Extremal Perturbations and Smooth Masks](https://mlsanity.com/paper/1910.08485/) (2019)
- [Guided Integrated Gradients: An Adaptive Path Method for Removing Noise](https://mlsanity.com/paper/2106.09788/) (2021)
- [Towards Better Explanations of Class Activation Mapping](https://mlsanity.com/paper/2102.05228/) (2021)
- [Visual Explanations via Iterated Integrated Attributions](https://mlsanity.com/paper/2310.18585/) (2023)
- [Interpreting Adversarial Examples by Activation Promotion and Suppression](https://mlsanity.com/paper/1904.02057/) (2019)
- [Empowering CAM-Based Methods with Capability to Generate Fine-Grained and High-Faithfulness Explanations](https://mlsanity.com/paper/2303.09171/) (2023)

## Abstract (arXiv)

Faithfulness metrics such as insertion and deletion evaluate how feature removal affects model outputs but overlook whether explanations preserve the computational pathway the network actually uses. We show that external metrics can be maximized through alternative pathways -- perturbations that reroute computation via different feature detectors while preserving output behavior. To address this, we propose activation preservation as a tractable proxy for preserving computational pathways We introduce Faithfulness-guided Ensemble Interpretation (FEI), which jointly optimizes external faithfulness (via ensemble quantile optimization of insertion/deletion curves) and internal faithfulness (via selective gradient clipping). Across VGG and ResNet on ImageNet and CUB-200-2011, FEI achieves state-of-the-art insertion/deletion scores while maintaining significantly lower activation deviation, showing that both external and internal faithfulness are essential for reliable explanations.