# "Why Should You Trust My Explanation?" Understanding Uncertainty in LIME Explanations

arXiv: https://arxiv.org/abs/1904.12991 · HTML: https://mlsanity.com/paper/1904.12991/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Apr 2019.

## Built on by (most cited first, 9 of 9)

- [Counterfactuals and Causability in Explainable Artificial Intelligence: Theory, Algorithms, and Applications](https://mlsanity.com/paper/2103.04244/) (2021)
- [A Survey on the Robustness of Feature Importance and Counterfactual Explanations](https://mlsanity.com/paper/2111.00358/) (2021)
- [Interpretation of Black Box NLP Models: A Survey](https://mlsanity.com/paper/2203.17081/) (2022)
- [Model Transparency and Interpretability : Survey and Application to the Insurance Industry](https://mlsanity.com/paper/2209.00562/) (2022)
- [Sampling Based On Natural Image Statistics Improves Local Surrogate Explainers](https://mlsanity.com/paper/2208.03961/) (2022)
- [Fairness of Explanations in Artificial Intelligence (AI): A Unifying Framework, Axioms, and Future Direction toward Responsible AI](https://mlsanity.com/paper/2605.09852/) (2026)
- [What's meant by explainable model: A Scoping Review](https://mlsanity.com/paper/2307.09673/) (2023)
- [Evaluating Explanatory Capabilities of Machine Learning Models in Medical Diagnostics: A Human-in-the-Loop Approach](https://mlsanity.com/paper/2403.19820/) (2024)
- [Automatic Detection of Deaths from Social Networking Sites](https://mlsanity.com/paper/2608.05183/) (2026)

## Abstract (arXiv)

Methods for interpreting machine learning black-box models increase the outcomes' transparency and in turn generates insight into the reliability and fairness of the algorithms. However, the interpretations themselves could contain significant uncertainty that undermines the trust in the outcomes and raises concern about the model's reliability. Focusing on the method "Local Interpretable Model-agnostic Explanations" (LIME), we demonstrate the presence of two sources of uncertainty, namely the randomness in its sampling procedure and the variation of interpretation quality across different input data points. Such uncertainty is present even in models with high training and test accuracy. We apply LIME to synthetic data and two public data sets, text classification in 20 Newsgroup and recidivism risk-scoring in COMPAS, to support our argument.