# Selective Explanations

arXiv: https://arxiv.org/abs/2405.19562 · HTML: https://mlsanity.com/paper/2405.19562/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 41 references that are in the map.
- First posted on arXiv 29 May 2024.

## Builds on (most cited first, 26 of 26)

- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- [Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles](https://mlsanity.com/paper/1612.01474/) (2016)
- [Character-level Convolutional Networks for Text Classification](https://mlsanity.com/paper/1509.01626/) (2015)
- [Visualizing and Understanding Convolutional Networks](https://mlsanity.com/paper/1311.2901/) (2013)
- [Selective Classification for Deep Neural Networks](https://mlsanity.com/paper/1705.08500/) (2017)
- [Captum: A unified and generic model interpretability library for PyTorch](https://mlsanity.com/paper/2009.07896/) (2020)
- [ToxiGen: A Large-Scale Machine-Generated Dataset for Adversarial and Implicit Hate Speech Detection](https://mlsanity.com/paper/2203.09509/) (2022)
- [SelectiveNet: A Deep Neural Network with an Integrated Reject Option](https://mlsanity.com/paper/1901.09192/) (2019)
- [Improving KernelSHAP: Practical Shapley Value Estimation via Linear Regression](https://mlsanity.com/paper/2012.01536/) (2020)
- [FastSHAP: Real-Time Shapley Value Estimation](https://mlsanity.com/paper/2107.07436/) (2021)
- [Sampling Permutations for Shapley Value Estimation](https://mlsanity.com/paper/2104.12199/) (2021)
- [CXPlain: Causal Explanations for Model Interpretation under Uncertainty](https://mlsanity.com/paper/1910.12336/) (2019)
- [Towards Human-centered Explainable AI: A Survey of User Studies for Model Explanations](https://mlsanity.com/paper/2210.11584/) (2022)
- [A Perspective on Explainable Artificial Intelligence Methods: SHAP and LIME](https://mlsanity.com/paper/2305.02012/) (2023)
- [Selective Classification via One-Sided Prediction](https://mlsanity.com/paper/2010.07853/) (2020)
- [Selective Regression Under Fairness Criteria](https://mlsanity.com/paper/2110.15403/) (2021)
- [Using Captum to Explain Generative Language Models](https://mlsanity.com/paper/2312.05491/) (2023)
- [Regression with reject option and application to kNN](https://mlsanity.com/paper/2006.16597/) (2020)
- [A Human-Grounded Evaluation of SHAP for Alert Processing](https://mlsanity.com/paper/1907.03324/) (2019)
- [Stochastic Amortization: A Unified Approach to Accelerate Feature and Data Attribution](https://mlsanity.com/paper/2401.15866/) (2024)
- [Selective Prediction via Training Dynamics](https://mlsanity.com/paper/2205.13532/) (2022)
- [CoRTX: Contrastive Framework for Real-time Explanation](https://mlsanity.com/paper/2303.02794/) (2023)
- [Multi-Level Explanations for Generative Language Models](https://mlsanity.com/paper/2403.14459/) (2024)
- [Efficient Explanations from Empirical Explainers](https://mlsanity.com/paper/2103.15429/) (2021)
- [LS-Tree: Model Interpretation When the Data Are Linguistic](https://mlsanity.com/paper/1902.04187/) (2019)
- [Efficient Shapley Values Estimation by Amortization for Text Classification](https://mlsanity.com/paper/2305.19998/) (2023)

## Built on by (most cited first, 2 of 2)

- [Local MDI+: Local Feature Importances for Tree-Based Models](https://mlsanity.com/paper/2506.08928/) (2025)
- [Transferring Visual Explainability of Self-Explaining Models to Prediction-Only Models without Additional Training](https://mlsanity.com/paper/2507.04380/) (2025)

## Abstract (arXiv)

Feature attribution methods explain black-box machine learning (ML) models by assigning importance scores to input features. These methods can be computationally expensive for large ML models. To address this challenge, there has been increasing efforts to develop amortized explainers, where a machine learning model is trained to predict feature attribution scores with only one inference. Despite their efficiency, amortized explainers can produce inaccurate predictions and misleading explanations. In this paper, we propose selective explanations, a novel feature attribution method that (i) detects when amortized explainers generate low-quality explanations and (ii) improves these explanations using a technique called explanations with initial guess. Our selective explanation method allows practitioners to specify the fraction of samples that receive explanations with initial guess, offering a principled way to bridge the gap between amortized explainers and their high-quality counterparts.