mlsanityData to 7 Oct 2026

XRAI: Better Attributions Through Regions

arXiv 1906.02825 ↗First posted 6 Jun 2019cs.CV stat.MLcited by 116

Not on the map. Why?

Abstract (arXiv)

Saliency methods can aid understanding of deep neural networks. Recent years have witnessed many improvements to saliency methods, as well as new ways for evaluating them. In this paper, we 1) present a novel region-based attribution method, XRAI, that builds upon integrated gradients (Sundararajan et al. 2017), 2) introduce evaluation methods for empirically assessing the quality of image-based saliency maps (Performance Information Curves (PICs)), and 3) contribute an axiom-based sanity check for attribution methods. Through empirical experiments and example results, we show that XRAI produces better results than other saliency methods for common models and the ImageNet dataset.

Key facts. Cited by 116 papers in the map (data through 7 Oct 2026). First posted on arXiv 6 Jun 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 19; 2022: 24; 2023: 22; 2024: 26; 2025: 13; 2026: 12.

116 papers.

Sort:

By area:

  1. 2021Towards Total Recall in Industrial Anomaly Detectioncited by 772
  2. 2021Benchmarking and Survey of Explanation Methods for Black Box Modelscited by 136
  3. 2022Algorithms to estimate Shapley value feature attributionscited by 114
  4. 2021Guided Integrated Gradients: An Adaptive Path Method for Removing Noisecited by 102
  5. 2021What I Cannot Predict, I Do Not Understand: A Human-Centered Evaluation Framework for Explainability Methodscited by 98
  6. 2022Impossibility Theorems for Feature Attributioncited by 78
  7. 2021Look at the Variance! Efficient Black-box Explanations with Sobol-based Sensitivity Analysiscited by 65
  8. 2024FakeInversion: Learning to Detect Images from Unseen Text-to-Image Models by Inverting Stable Diffusioncited by 61
  9. 2021Group-CAM: Group Score-Weighted Visual Explanations for Deep Convolutional Networkscited by 36
  10. 2024Gradient based Feature Attribution in Explainable AI: A Technical Reviewcited by 34
  11. 2022Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measurecited by 33
  12. 2024LeGrad: An Explainability Method for Vision Transformers via Feature Formation Sensitivitycited by 32
38 more
  1. 2021CAMERAS: Enhanced Resolution And Sanity preserving Class Activation Mapping for image saliencycited by 28
  2. 2024Decoupling Pixel Flipping and Occlusion Strategy for Consistent XAI Benchmarkscited by 28
  3. 2021Best of both worlds: local and global explanations with human-understandable conceptscited by 22
  4. 2023GLIME: General, Stable and Local LIME Explanationcited by 21
  5. 2023CoRTX: Contrastive Framework for Real-time Explanationcited by 20
  6. 2022Evaluating Feature Attribution Methods in the Image Domaincited by 19
  7. 2022Quantifying the Knowledge in a DNN to Explain Knowledge Distillation for Classificationcited by 19
  8. 2023Visual Explanations via Iterated Integrated Attributionscited by 18
  9. 2022Attri-VAE: attribute-based interpretable representations of medical images with variational autoencoderscited by 17
  10. 2021A Game-Theoretic Taxonomy of Visual Concepts in DNNscited by 16
  11. 2022Attribution-based XAI Methods in Computer Vision: A Reviewcited by 16
  12. 2023IDGI: A Framework to Eliminate Explanation Noise from Integrated Gradientscited by 16
  13. 2023Towards Better Explanations for Object Detectioncited by 14
  14. 2021Building Reliable Explanations of Unreliable Neural Networks: Locally Smoothing Perspective of Model Interpretationcited by 12
  15. 2024IG2: Integrated Gradient on Iterative Gradient Path for Feature Attributioncited by 12
  16. 2022Saliency Cards: A Framework to Characterize and Compare Saliency Methodscited by 11
  17. 2024Explainable AI (XAI) in Image Segmentation in Medicine, Industry, and Beyond: A Surveycited by 11
  18. 2024Data Augmentation for Image Classification using Generative AIcited by 11
  19. 2022Predicting and Explaining Mobile UI Tappability with Vision Modeling and Saliency Analysiscited by 9
  20. 2023AUTOLYCUS: Exploiting Explainable AI (XAI) for Model Extraction Attacks against Interpretable Modelscited by 9
  21. 2023Feature Perturbation Augmentation for Reliable Evaluation of Importance Estimators in Neural Networkscited by 9
  22. 2023Integrated Decision Gradients: Compute Your Attributions Where the Model Makes Its Decisioncited by 9
  23. 2023MAEA: Multimodal Attribution for Embodied AIcited by 9
  24. 2023Deep Integrated Explanationscited by 9
  25. 2025Attention, Please! PixelSHAP Reveals What Vision-Language Models Actually Focus Oncited by 9
  26. 2021Attribution of Predictive Uncertainties in Classification Modelscited by 8
  27. 2022This changes to that : Combining causal and non-causal explanations to generate disease progression in capsule endoscopycited by 8
  28. 2024Beyond Pixels: Enhancing LIME with Hierarchical Features and Segmentation Foundation Modelscited by 8
  29. 2022FD-CAM: Improving Faithfulness and Discriminability of Visual Explanation for CNNscited by 7
  30. 2023Negative Flux Aggregation to Estimate Feature Attributionscited by 7
  31. 2023G-CAME: Gaussian-Class Activation Mapping Explainer for Object Detectorscited by 7
  32. 2023An adversarial attack approach for eXplainable AI evaluation on deepfake detection modelscited by 7
  33. 2023MFABA: A More Faithful and Accelerated Boundary-based Attribution Method for Deep Neural Networkscited by 7
  34. 2024Abstraction Alignment: Comparing Model-Learned and Human-Encoded Conceptual Relationshipscited by 7
  35. 2024Efficient and Concise Explanations for Object Detection with Gaussian-Class Activation Mapping Explainercited by 6
  36. 2024How explainable AI affects human performance: A systematic review of the behavioural consequences of saliency mapscited by 6
  37. 2025Evaluating Model Explanations without Ground Truthcited by 6
  38. 2021Improving Attribution Methods by Learning Submodular Functionscited by 5

50 most cited of 116 All 116 (JSON)