# When Explanations Lie: Why Many Modified BP Attributions Fail

arXiv: https://arxiv.org/abs/1912.09818 · HTML: https://mlsanity.com/paper/1912.09818/

## Key facts

- Cited by 92 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Dec 2019.

## Built on by (most cited first, 30 of 92)

- [From Anecdotal Evidence to Quantitative Evaluation Methods: A Systematic Review on Evaluating Explainable AI](https://mlsanity.com/paper/2201.08164/) (2022)
- [CRAFT: Concept Recursive Activation FacTorization for Explainability](https://mlsanity.com/paper/2211.10154/) (2022)
- [Quantus: An Explainable AI Toolkit for Responsible Evaluation of Neural Network Explanations and Beyond](https://mlsanity.com/paper/2202.06861/) (2022)
- [A Holistic Approach to Unifying Automatic Concept Extraction and Concept Importance Estimation](https://mlsanity.com/paper/2306.07304/) (2023)
- [Towards a Science of Human-AI Decision Making: A Survey of Empirical Studies](https://mlsanity.com/paper/2112.11471/) (2021)
- [A Primer on the Inner Workings of Transformer-based Language Models](https://mlsanity.com/paper/2405.00208/) (2024)
- [XProtoNet: Diagnosis in Chest Radiography with Global and Local Explanations](https://mlsanity.com/paper/2103.10663/) (2021)
- [GlanceNets: Interpretabile, Leak-proof Concept-based Models](https://mlsanity.com/paper/2205.15612/) (2022)
- [Archetypal SAE: Adaptive and Stable Dictionary Learning for Concept Extraction in Large Vision Models](https://mlsanity.com/paper/2502.12892/) (2025)
- [Universal Sparse Autoencoders: Interpretable Cross-Model Concept Alignment](https://mlsanity.com/paper/2502.03714/) (2025)
- [Leveraging Explanations in Interactive Machine Learning: An Overview](https://mlsanity.com/paper/2207.14526/) (2022)
- [Do Input Gradients Highlight Discriminative Features?](https://mlsanity.com/paper/2102.12781/) (2021)
- [Concept-level Debugging of Part-Prototype Networks](https://mlsanity.com/paper/2205.15769/) (2022)
- [Logic-Based Explainability in Machine Learning](https://mlsanity.com/paper/2211.00541/) (2022)
- [This looks more like that: Enhancing Self-Explaining Models by Prototypical Relevance Propagation](https://mlsanity.com/paper/2108.12204/) (2021)
- [Rethinking Attention-Model Explainability through Faithfulness Violation Test](https://mlsanity.com/paper/2201.12114/) (2022)
- [Explaining a Series of Models by Propagating Shapley Values](https://mlsanity.com/paper/2105.00108/) (2021)
- [The Meta-Evaluation Problem in Explainable AI: Identifying Reliable Estimators with MetaQuantus](https://mlsanity.com/paper/2302.07265/) (2023)
- [Neural Response Interpretation through the Lens of Critical Pathways](https://mlsanity.com/paper/2103.16886/) (2021)
- [Finding the right XAI method -- A Guide for the Evaluation and Ranking of Explainable AI Methods in Climate Science](https://mlsanity.com/paper/2303.00652/) (2023)
- [Unlocking Feature Visualization for Deeper Networks with MAgnitude Constrained Optimization](https://mlsanity.com/paper/2306.06805/) (2023)
- [Xplique: A Deep Learning Explainability Toolbox](https://mlsanity.com/paper/2206.04394/) (2022)
- [Shortcomings of Top-Down Randomization-Based Sanity Checks for Evaluations of Deep Neural Network Explanations](https://mlsanity.com/paper/2211.12486/) (2022)
- [Do Explanations Explain? Model Knows Best](https://mlsanity.com/paper/2203.02269/) (2022)
- [Scrutinizing XAI using linear ground-truth data with suppressor variables](https://mlsanity.com/paper/2111.07473/) (2021)
- [On Computing Probabilistic Abductive Explanations](https://mlsanity.com/paper/2212.05990/) (2022)
- [Fine-Grained Neural Network Explanation by Identifying Input Features with Predictive Information](https://mlsanity.com/paper/2110.01471/) (2021)
- [Right for the Wrong Reason: Can Interpretable ML Techniques Detect Spurious Correlations?](https://mlsanity.com/paper/2307.12344/) (2023)
- [A Fine-grained Interpretability Evaluation Benchmark for Neural NLP](https://mlsanity.com/paper/2205.11097/) (2022)
- [Why and How LLMs Hallucinate: Connecting the Dots with Subsequence Associations](https://mlsanity.com/paper/2504.12691/) (2025)

## Abstract (arXiv)

Attribution methods aim to explain a neural network's prediction by highlighting the most relevant image areas. A popular approach is to backpropagate (BP) a custom relevance score using modified rules, rather than the gradient. We analyze an extensive set of modified BP methods: Deep Taylor Decomposition, Layer-wise Relevance Propagation (LRP), Excitation BP, PatternAttribution, DeepLIFT, Deconv, RectGrad, and Guided BP. We find empirically that the explanations of all mentioned methods, except for DeepLIFT, are independent of the parameters of later layers. We provide theoretical insights for this surprising behavior and also analyze why DeepLIFT does not suffer from this limitation. Empirically, we measure how information of later layers is ignored by using our new metric, cosine similarity convergence (CSC). The paper provides a framework to assess the faithfulness of new and existing modified BP methods theoretically and empirically. For code see: https://github.com/berleon/when-explanations-lie