mlsanityData to 7 Oct 2026

Restricting the Flow: Information Bottlenecks for Attribution

arXiv 2001.00396 ↗First posted 2 Jan 2020stat.ML cs.CV cs.LGcited by 110

Not on the map. Why?

Abstract (arXiv)

Attribution methods provide insights into the decision-making of machine learning models like artificial neural networks. For a given input sample, they assign a relevance score to each individual input variable, such as the pixels of an image. In this work we adapt the information bottleneck concept for attribution. By adding noise to intermediate feature maps we restrict the flow of information and can quantify (in bits) how much information image regions provide. We compare our method against ten baselines using three different metrics on VGG-16 and ResNet-50, and find that our methods outperform all baselines in five out of six settings. The method's information-theoretic foundation provides an absolute frame of reference for attribution values (bits) and a guarantee that regions scored close to zero are not necessary for the network's decision. For reviews: https://openreview.net/forum?id=S1xWh1rYwB For code: https://github.com/BioroboticsLab/IBA

Key facts. Cited by 110 papers in the map (data through 7 Oct 2026). First posted on arXiv 2 Jan 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 28; 2022: 19; 2023: 19; 2024: 16; 2025: 18; 2026: 10.

110 papers.

Sort:

By area:

  1. 2021FastSHAP: Real-Time Shapley Value Estimationcited by 123
  2. 2021Acquisition of Chess Knowledge in AlphaZerocited by 120
  3. 2021BBAM: Bounding Box Attribution Map for Weakly Supervised Semantic and Instance Segmentationcited by 104
  4. 2021Reducing Information Bottleneck for Weakly Supervised Semantic Segmentationcited by 65
  5. 2023Task-specific Fine-tuning via Variational Information Bottleneck for Weakly-supervised Pathology Whole Slide Image Classificationcited by 54
  6. 2023Learning Bottleneck Concepts in Image Classificationcited by 49
  7. 2021Improving Subgraph Recognition with Variational Graph Information Bottleneckcited by 45
  8. 2023Visual Explanations of Image-Text Representations via Multi-Modal Information Bottleneck Attributioncited by 34
  9. 2021Neural Response Interpretation through the Lens of Critical Pathwayscited by 32
  10. 2022Distilling Robust and Non-Robust Features in Adversarial Examples by Information Bottleneckcited by 30
  11. 2022Towards Better Understanding Attribution Methodscited by 28
  12. 2021The Eval4NLP Shared Task on Explainable Quality Estimation: Overview and Resultscited by 27
38 more
  1. 2023AD-KD: Attribution-Driven Knowledge Distillation for Language Model Compressioncited by 26
  2. 2023Opti-CAM: Optimizing saliency maps for interpretabilitycited by 25
  3. 2022Listen to Interpret: Post-hoc Interpretability for Audio Networks with NMFcited by 23
  4. 2021Translation Error Detection as Rationale Extractioncited by 22
  5. 2022REV: Information-Theoretic Evaluation of Free-Text Rationalescited by 22
  6. 2024Optimal ablation for interpretabilitycited by 21
  7. 2022Do Explanations Explain? Model Knows Bestcited by 20
  8. 2025A Comprehensive Survey on Self-Interpretable Neural Networkscited by 18
  9. 2021Fine-Grained Neural Network Explanation by Identifying Input Features with Predictive Informationcited by 17
  10. 2021Integrated Grad-CAM: Sensitivity-Aware Visual Explanation of Deep Convolutional Networks via Integrated Gradient-Based Scoringcited by 16
  11. 2021A Game-Theoretic Taxonomy of Visual Concepts in DNNscited by 16
  12. 2023HarsanyiNet: Computing Accurate Shapley Values in a Single Forward Propagationcited by 16
  13. 2021Drop-Bottleneck: Learning Discrete Compressed Representation for Noise-Robust Explorationcited by 15
  14. 2022Towards Explainable Evaluation Metrics for Natural Language Generationcited by 15
  15. 2024DEAL: Disentangle and Localize Concept-level Explanations for VLMscited by 14
  16. 2021Explaining Neural Network Predictions on Sentence Pairs via Learning Word-Group Maskscited by 12
  17. 2021RELAX: Representation Learning Explainabilitycited by 12
  18. 2022Comparing the Decision-Making Mechanisms by Transformers and CNNs via Explanation Methodscited by 12
  19. 2021Ada-SISE: Adaptive Semantic Input Sampling for Efficient Explanation of Convolutional Neural Networkscited by 11
  20. 2023Understanding and Unifying Fourteen Attribution Methods with Taylor Interactionscited by 11
  21. 2022New Definitions and Evaluations for Saliency Methods: Staying Intrinsic, Complete and Soundcited by 10
  22. 2023Towards Trust of Explainable AI in Thyroid Nodule Diagnosiscited by 10
  23. 2021Efficient Explanations from Empirical Explainerscited by 9
  24. 2021Explaining COVID-19 and Thoracic Pathology Model Predictions by Identifying Informative Input Featurescited by 9
  25. 2023Pixel-Level Explanation of Multiple Instance Learning Models in Biomedical Single Cell Imagescited by 9
  26. 2022Interpretability with full complexity by constraining feature informationcited by 7
  27. 2024XEdgeAI: A Human-centered Industrial Inspection Framework with Data-centric Explainable Edge AI Approachcited by 7
  28. 2026CRAFT: Adapting VLA Models to Contact-rich Manipulation via Force-aware Curriculum Fine-tuningcited by 7
  29. 2021Information Bottleneck Approach to Spatial Attention Learningcited by 6
  30. 2021Making a Bird AI Expert Work for You and Mecited by 6
  31. 2023Improving Interpretability via Explicit Word Interaction Graph Layercited by 6
  32. 2023Prototypical Self-Explainable Models Without Re-trainingcited by 6
  33. 2025Comprehensive Information Bottleneck for Unveiling Universal Attribution to Interpret Vision Transformerscited by 6
  34. 2022Reducing Hallucinations in Neural Machine Translation with Feature Attributioncited by 5
  35. 2023Tackling Interpretability in Audio Classification Networks with Non-negative Matrix Factorizationcited by 5
  36. 2025Class-Dependent Perturbation Effects in Evaluating Time Series Attributionscited by 5
  37. 2025DocVXQA: Context-Aware Visual Explanations for Document Question Answeringcited by 5
  38. 2021Benchmarking Perturbation-based Saliency Maps for Explaining Atari Agentscited by 4

50 most cited of 110 All 110 (JSON)