# Position: Explanation Stability Is a Property of the Model Method Pair, Not the Model

arXiv: https://arxiv.org/abs/2607.16652 · HTML: https://mlsanity.com/paper/2607.16652/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 36 references that are in the map.
- First posted on arXiv 18 Jul 2026.

## Builds on (most cited first, 24 of 24)

- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- [Identity Mappings in Deep Residual Networks](https://mlsanity.com/paper/1603.05027/) (2016)
- [Sanity Checks for Saliency Maps](https://mlsanity.com/paper/1810.03292/) (2018)
- [Visualizing the Loss Landscape of Neural Nets](https://mlsanity.com/paper/1712.09913/) (2017)
- [Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networks](https://mlsanity.com/paper/1710.11063/) (2017)
- [Big Transfer (BiT): General Visual Representation Learning](https://mlsanity.com/paper/1912.11370/) (2019)
- [Do Vision Transformers See Like Convolutional Neural Networks?](https://mlsanity.com/paper/2108.08810/) (2021)
- [A Benchmark for Interpretability Methods in Deep Neural Networks](https://mlsanity.com/paper/1806.10758/) (2018)
- [Interpretation of Neural Networks is Fragile](https://mlsanity.com/paper/1710.10547/) (2017)
- [Transfusion: Understanding Transfer Learning for Medical Imaging](https://mlsanity.com/paper/1902.07208/) (2019)
- [The (Un)reliability of saliency methods](https://mlsanity.com/paper/1711.00867/) (2017)
- [From Anecdotal Evidence to Quantitative Evaluation Methods: A Systematic Review on Evaluating Explainable AI](https://mlsanity.com/paper/2201.08164/) (2022)
- [Quantus: An Explainable AI Toolkit for Responsible Evaluation of Neural Network Explanations and Beyond](https://mlsanity.com/paper/2202.06861/) (2022)
- [The Disagreement Problem in Explainable Machine Learning: A Practitioner's Perspective](https://mlsanity.com/paper/2202.01602/) (2022)
- [OpenXAI: Towards a Transparent Evaluation of Model Explanations](https://mlsanity.com/paper/2206.11104/) (2022)
- [Which Explanation Should I Choose? A Function Approximation Perspective to Characterizing Post Hoc Explanations](https://mlsanity.com/paper/2206.01254/) (2022)
- [Evaluation Gaps in Machine Learning Practice](https://mlsanity.com/paper/2205.05256/) (2022)
- [The Meta-Evaluation Problem in Explainable AI: Identifying Reliable Estimators with MetaQuantus](https://mlsanity.com/paper/2302.07265/) (2023)
- [Benchmarking XAI Explanations with Human-Aligned Evaluations](https://mlsanity.com/paper/2411.02470/) (2024)
- [Reckoning with the Disagreement Problem: Explanation Consensus as a Training Objective](https://mlsanity.com/paper/2303.13299/) (2023)
- [On the Evaluation Consistency of Attribution-based Explanations](https://mlsanity.com/paper/2407.19471/) (2024)
- [Evaluating Explainability: A Framework for Systematic Assessment and Reporting of Explainable AI Features](https://mlsanity.com/paper/2506.13917/) (2025)

## Abstract (arXiv)

This position paper argues that claims about explanation stability are scientifically invalid without cross method validation. Just as statistical significance requires the test statistic to be specified, stability should either be evaluated across multiple attribution paradigms or explicitly scoped to the computational objective of a single method. In controlled chest X ray experiments, DenseNet201, ResNet50V2, and InceptionV3 achieved AUC values above 99%, yet their stability rankings reversed across attribution methods. LayerCAM ranked InceptionV3 as the most stable model, with an IoU of 0.777, whereas GradCAM++ favored DenseNet201 and reduced InceptionV3 stability score by 17.3%. These findings demonstrate that explanation stability is an emergent property of the model method pair rather than an intrinsic characteristic of the model alone. We therefore argue that explanation based claims should be validated across multiple attribution methods and that regulatory submissions should explicitly specify the attribution operators used to avoid creating illusory safety assurances.