mlsanityData to 7 Oct 2026

Explain Yourself, Briefly! Self-Explaining Neural Networks with Concise Sufficient Reasons

arXiv 2502.03391 ↗First posted 5 Feb 2025cs.LG cs.LOcited by 11

Not on the map. Why?

Abstract (arXiv)

*Minimal sufficient reasons* represent a prevalent form of explanation - the smallest subset of input features which, when held constant at their corresponding values, ensure that the prediction remains unchanged. Previous *post-hoc* methods attempt to obtain such explanations but face two main limitations: (1) Obtaining these subsets poses a computational challenge, leading most scalable methods to converge towards suboptimal, less meaningful subsets; (2) These methods heavily rely on sampling out-of-distribution input assignments, potentially resulting in counterintuitive behaviors. To tackle these limitations, we propose in this work a self-supervised training approach, which we term *sufficient subset training* (SST). Using SST, we train models to generate concise sufficient reasons for their predictions as an integral part of their output. Our results indicate that our framework produces succinct and faithful subsets substantially more efficiently than competing post-hoc methods, while maintaining comparable predictive performance.

Key facts. Cited by 11 papers in the map (data through 7 Oct 2026). Builds on 79 of its 102 references that are in the map. First posted on arXiv 5 Feb 2025.

Builds on

79 of 102 references.

References · 79

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  3. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  4. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  5. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,040
  6. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  7. 2015A large annotated corpus for learning natural language inferencecited by 1,837
  8. 2020Concept Bottleneck Modelscited by 1,022
  9. 2017Reluplex: An Efficient SMT Solver for Verifying Deep Neural Networkscited by 739
  10. 2018This Looks Like That: Deep Learning for Interpretable Image Recognitioncited by 731
  11. 2019Adversarial Training for Free!cited by 638
  12. 2017Interpretable Explanations of Black Boxes by Meaningful Perturbationcited by 611
38 more
  1. 2019ERASER: A Benchmark to Evaluate Rationalized NLP Modelscited by 523
  2. 2018Towards Robust Interpretability with Self-Explaining Neural Networkscited by 505
  3. 2019Fooling LIME and SHAP: Adversarial Attacks on Post hoc Explanation Methodscited by 440
  4. 2019The many Shapley values for model explanationcited by 341
  5. 2020Neural Additive Models: Interpretable Machine Learning with Neural Netscited by 318
  6. 2019On the (In)fidelity and Sensitivity for Explanationscited by 309
  7. 2021Beta-CROWN: Efficient Bound Propagation with Per-neuron Split Constraints for Complete and Incomplete Neural Network Robustness Verificationcited by 224
  8. 2020Unsupervised Commonsense Question Answering with Self-Talkcited by 161
  9. 2018Abduction-Based Explanations for Machine Learning Modelscited by 159
  10. 2018A Symbolic Approach to Explaining Bayesian Network Classifierscited by 130
  11. 2020Learning to Faithfully Rationalize by Constructioncited by 113
  12. 2023Probabilistic Concept Bottleneck Modelscited by 101
  13. 2020On The Reasons Behind Decisionscited by 99
  14. 2021ProtGNN: Towards Self-Explaining Graph Neural Networkscited by 96
  15. 2021The Out-of-Distribution Problem in Explainability and Search Methods for Feature Importance Explanationscited by 86
  16. 2019LassoNet: A Neural Network with Feature Sparsitycited by 81
  17. 2024Marabou 2.0: A Versatile Formal Analyzer of Neural Networkscited by 81
  18. 2020Model Interpretability through the Lens of Computational Complexitycited by 75
  19. 2022GlanceNets: Interpretabile, Leak-proof Concept-based Modelscited by 73
  20. 2021Improving Deep Learning Interpretability by Saliency Guided Trainingcited by 68
  21. 2021Have We Learned to Explain?: How Interpretability Methods Can Learn to Encode Predictions in their Interpretationscited by 59
  22. 2021On Explaining Random Forests with SATcited by 57
  23. 2020Explaining Naive Bayes and Other Linear Classifiers with Polynomial Time and Delaycited by 56
  24. 2021On Guaranteed Optimal Robust Explanations for NLP Modelscited by 54
  25. 2020On Explaining Decision Treescited by 49
  26. 2022Logic-Based Explainability in Machine Learningcited by 47
  27. 2021Local Explanations via Necessity and Sufficiency: Unifying Theory and Practicecited by 45
  28. 2021SelfExplain: A Self-Explaining Architecture for Neural Text Classifierscited by 44
  29. 2022Towards Formal XAI: Formally Approximate Minimal Explanations of Neural Networkscited by 42
  30. 2021On the Computational Intelligibility of Boolean Classifierscited by 40
  31. 2021Explanations for Monotonic Classifierscited by 39
  32. 2022Don't Lie to Me! Robust and Efficient Explainability with Verified Perturbation Analysiscited by 38
  33. 2021On Efficiently Explaining Graph-Based Classifierscited by 37
  34. 2022On Computing Probabilistic Explanations for Decision Treescited by 35
  35. 2021Locally Sparse Neural Networks for Tabular Biomedical Datacited by 34
  36. 2021Trading Complexity for Sparsity in Random Forest Explanationscited by 32
  37. 2021Provably efficient, succinct, and precise explanationscited by 32
  38. 2018What made you do this? Understanding black-box decisions with sufficient input subsetscited by 31

50 most cited of 79 All 79 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 5; 2026: 6.

11 papers.

  1. 2025Explaining, Fast and Slow: Abstraction and Refinement of Provable Explanationscited by 13
  2. 2026Formal Mechanistic Interpretability: Automated Circuit Discovery with Provable Guaranteescited by 10
  3. 2025SHAP Meets Tensor Networks: Provably Tractable Explanations with Parallelismcited by 8
  4. 2025Self-Explaining Neural Networks for Business Process Monitoringcited by 5
  5. 2025Additive Models Explained: A Computational Complexity Approachcited by 3
  6. 2026GNN Explanations that do not Explain and How to find Themcited by 3
  7. 2026FAME: Formal Abstract Minimal Explanation for Neural Networkscited by 3
  8. 2026Provably Explaining Neural Additive Modelscited by 1
  9. 2026Unifying Formal Explanations: A Complexity-Theoretic Perspectivecited by 1
  10. 2025Keypoint Counting Classifiers: Turning Vision Transformers into Self-Explainable Models Without Trainingcited by 0
  11. 2026The Sufficiency-Conciseness Trade-off in LLM Self-Explanation from an Information Bottleneck Perspectivecited by 0