mlsanityData to 7 Oct 2026

Sanity Simulations for Saliency Methods

arXiv 2105.06506 ↗First posted 13 May 2021cs.LGcited by 15

Not on the map. Why?

Abstract (arXiv)

Saliency methods are a popular class of feature attribution explanation methods that aim to capture a model's predictive reasoning by identifying "important" pixels in an input image. However, the development and adoption of these methods are hindered by the lack of access to ground-truth model reasoning, which prevents accurate evaluation. In this work, we design a synthetic benchmarking framework, SMERF, that allows us to perform ground-truth-based evaluation while controlling the complexity of the model's reasoning. Experimentally, SMERF reveals significant limitations in existing saliency methods and, as a result, represents a useful tool for the development of new saliency methods.

Key facts. Cited by 15 papers in the map (data through 7 Oct 2026). Builds on 29 of its 37 references that are in the map. First posted on arXiv 13 May 2021.

Builds on

29 of 37 references.

References · 29

  1. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  2. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  3. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  4. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  5. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  6. 2018CBAM: Convolutional Block Attention Modulecited by 2,463
  7. 2015Learning Deep Features for Discriminative Localizationcited by 2,430
  8. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,348
  9. 2013Visualizing and Understanding Convolutional Networkscited by 2,278
  10. 2017Learning Important Features Through Propagating Activation Differencescited by 1,454
  11. 2018Sanity Checks for Saliency Mapscited by 1,149
  12. 2017SmoothGrad: removing noise by adding noisecited by 1,097
17 more
  1. 2017Counterfactual Explanations without Opening the Black Box: Automated Decisions and the GDPRcited by 1,086
  2. 2014Striving for Simplicity: The All Convolutional Netcited by 983
  3. 2017Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networkscited by 936
  4. 2019Score-CAM: Score-Weighted Visual Explanations for Convolutional Neural Networkscited by 575
  5. 2018Towards Robust Interpretability with Self-Explaining Neural Networkscited by 505
  6. 2019Res2Net: A New Multi-scale Backbone Architecturecited by 468
  7. 2018A Benchmark for Interpretability Methods in Deep Neural Networkscited by 467
  8. 2018Pseudo-LiDAR from Visual Depth Estimation: Bridging the Gap in 3D Object Detection for Autonomous Drivingcited by 455
  9. 2017The (Un)reliability of saliency methodscited by 331
  10. 2016Top-down Neural Attention by Excitation Backpropcited by 318
  11. 2020Debugging Tests for Model Explanationscited by 146
  12. 2019Sanity Checks for Saliency Metricscited by 119
  13. 2021Do Feature Attribution Methods Correctly Attribute Features?cited by 98
  14. 2019Benchmarking Attribution Methods with Relative Feature Importancecited by 89
  15. 2018iNNvestigate neural networks!cited by 58
  16. 2021Do Input Gradients Highlight Discriminative Features?cited by 52
  17. 2020Assessing the (Un)Trustworthiness of Saliency Maps for Localizing Abnormalities in Medical Imagingcited by 32

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 6; 2023: 3; 2024: 2; 2025: 3; 2026: 1.

15 papers.

Sort:

  1. 2022OpenXAI: Towards a Transparent Evaluation of Model Explanationscited by 125
  2. 2022Post hoc Explanations may be Ineffective for Detecting Unknown Spurious Correlationcited by 59
  3. 2022Use-Case-Grounded Simulations for Explanation Evaluationcited by 19
  4. 2023Studying How to Efficiently and Effectively Guide Models with Explanationscited by 16
  5. 2022Assessing Out-of-Domain Language Model Performance from Few Examplescited by 6
  6. 2022Towards a More Rigorous Science of Blindspot Discovery in Image Classification Modelscited by 5
  7. 2024Toward Understanding the Disagreement Problem in Neural Network Feature Attributioncited by 5
  8. 2022Responsibility: An Example-based Explainable AI approach via Training Process Inspectioncited by 2
  9. 2023Sanity Checks for Saliency Methods Explaining Object Detectorscited by 2
  10. 2025Time series saliency maps: explaining models across multiple domainscited by 2
  11. 2024CNN-based explanation ensembling for dataset, representation and explanations evaluationcited by 1
  12. 2025FaCT: Faithful Concept Traces for Explaining Neural Network Decisionscited by 1
3 more
  1. 2026Exploring SAIG Methods for an Objective Evaluation of XAIcited by 1
  2. 2023Explaining Deep Face Algorithms through Visualization: A Surveycited by 0
  3. 2025TriGuard: Testing Model Safety with Attribution Entropy, Verification, and Driftcited by 0