mlsanityData to 14 Oct 2026

Causally Reliable Concept Bottleneck Models

arXiv 2503.04363 ↗First posted 6 Mar 2025cs.LG cs.AIcited by 17

Not on the map. Why?

Abstract (arXiv)

Concept-based models are an emerging paradigm in deep learning that constrains the inference process to operate through human-interpretable variables, facilitating explainability and human interaction. However, these architectures, on par with popular opaque neural models, fail to account for the true causal mechanisms underlying the target phenomena represented in the data. This hampers their ability to support causal reasoning tasks, limits out-of-distribution generalization, and hinders the implementation of fairness constraints. To overcome these issues, we propose Causally reliable Concept Bottleneck Models (C2BMs), a class of concept-based architectures that enforce reasoning through a bottleneck of concepts structured according to a model of the real-world causal mechanisms. We also introduce a pipeline to automatically learn this structure from observational data and unstructured background knowledge (e.g., scientific literature). Experimental evidence suggests that C2BMs are more interpretable, causally reliable, and improve responsiveness to interventions w.r.t. standard opaque and concept-based models, while maintaining their accuracy.

Key facts. Cited by 17 papers in the map (data through 14 Oct 2026). Builds on 57 of its 93 references that are in the map. First posted on arXiv 6 Mar 2025.

Builds on

57 of 93 references.

References · 57

  1. 2017Attention Is All You Needcited by 60,361
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  3. 2020Language Models are Few-Shot Learnerscited by 32,419
  4. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  5. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  6. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  7. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,708
  8. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,070
  9. 2014Deep Learning Face Attributes in the Wildcited by 3,693
  10. 2023Retrieval-Augmented Generation for Large Language Models: A Surveycited by 2,212
  11. 2022Survey of Hallucination in Natural Language Generationcited by 2,075
  12. 2024Mixtral of Expertscited by 1,823
38 more
  1. 2019CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparisoncited by 1,423
  2. 2017Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)cited by 1,319
  3. 2020Concept Bottleneck Modelscited by 1,032
  4. 2016HyperNetworkscited by 1,027
  5. 2017ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseasescited by 972
  6. 2018This Looks Like That: Deep Learning for Interpretable Image Recognitioncited by 737
  7. 2023Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Modelscited by 698
  8. 2018Towards Robust Interpretability with Self-Explaining Neural Networkscited by 505
  9. 2022Language in a Bottle: Language Model Guided Concept Bottlenecks for Interpretable Image Classificationcited by 312
  10. 2022Post-hoc Concept Bottleneck Modelscited by 293
  11. 2023Label-Free Concept Bottleneck Modelscited by 285
  12. 2020Concept Whitening for Interpretable Image Recognitioncited by 225
  13. 2021D'ya like DAGs? A Survey on Structure Learning and Causal Discoverycited by 211
  14. 2013Learning Bayesian Networks: The Combination of Knowledge and Statistical Datacited by 187
  15. 2022CRAFT: Concept Recursive Activation FacTorization for Explainabilitycited by 178
  16. 2022From Attribution Maps to Human-Understandable Explanations through Concept Relevance Propagationcited by 141
  17. 2019Explaining Classifiers with Causal Concept Effect (CaCE)cited by 135
  18. 2019Learning Neural Causal Models from Unknown Interventionscited by 113
  19. 2022Causal Machine Learning: A Survey and Open Problemscited by 110
  20. 2023Causal-learn: Causal Discovery in Pythoncited by 109
  21. 2023Probabilistic Concept Bottleneck Modelscited by 101
  22. 2023Concept-based Explainable Artificial Intelligence: A Surveycited by 98
  23. 2023CXR-CLIP: Toward Large Scale Chest X-ray Language-Image Pre-trainingcited by 82
  24. 2023Can large language models build causal graphs?cited by 73
  25. 2013Counterfactual Probabilities: Computational Methods, Bounds and Applicationscited by 67
  26. 2024VLG-CBM: Training Concept Bottleneck Models with Vision-Language Guidancecited by 65
  27. 2023A Survey on Causal Discovery: Theory and Practicecited by 63
  28. 2022Concept Embedding Models: Beyond the Accuracy-Explainability Trade-Offcited by 61
  29. 2023Interpretable Neural-Symbolic Concept Reasoningcited by 57
  30. 2024Stochastic Concept Bottleneck Modelscited by 51
  31. 2022Out-of-distribution Generalization with Causal Invariant Transformationscited by 45
  32. 2023Learning to Receive Help: Intervention-Aware Concept Embedding Modelscited by 41
  33. 2023Investigating the Effectiveness of Task-Agnostic Prefix Prompt for Instruction Followingcited by 33
  34. 2023Learning to Intervene on Concept Bottleneckscited by 29
  35. 2024Causal Concept Graph Models: Beyond Causal Opacity in Deep Learningcited by 29
  36. 2017Fine-grained Visual-textual Representation Learningcited by 27
  37. 2020Structural Causal Models Are (Solvable by) Credal Networkscited by 22
  38. 2024Neural Causal Abstractionscited by 21

50 most cited of 57 All 57 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 5; 2026: 12.

17 papers.

Sort:

  1. 2025If Concept Bottlenecks are the Question, are Foundation Models the Answer?cited by 15
  2. 2025Graph Concept Bottleneck Modelscited by 9
  3. 2025V-CEM: Bridging Performance and Intervenability in Concept-based Modelscited by 3
  4. 2025Towards Reasonable Concept Bottleneck Modelscited by 3
  5. 2026Mixture of Concept Bottleneck Expertscited by 2
  6. 2026Matryoshka Concept Bottleneck Modelscited by 2
  7. 2026Radiologist-Guided Causal Concept Bottleneck Models for Chest X-Ray Interpretationcited by 1
  8. 2025Neural Concept Verifier: Scaling Prover-Verifier Games via Concept Encodingscited by 0
  9. 2026Enhanced Data-Driven Product Development via Gradient Based Optimization and Conformalized Monte Carlo Dropout Uncertainty Estimationcited by 0
  10. 2026Interpretability in Deep Time Series Models Demands Semantic Alignmentcited by 0
  11. 2026Federated Concept-Based Models: Interpretable models with distributed supervisioncited by 0
  12. 2026Causal Neural Probabilistic Circuitscited by 0
5 more
  1. 2026Hyperbolic Concept Bottleneck Modelscited by 0
  2. 2026Measuring What Matters: Synthetic Benchmarks for Concept Bottleneck Modelscited by 0
  3. 2026Formal Concept Lattices are Good Semantic Scaffolds for Concept-Based Learningcited by 0
  4. 2026TRACE: A Concept Bottleneck Model for Longitudinal 3D Glioblastoma Response Assessmentcited by 0
  5. 2026Beyond Heatmaps: Unsupervised Concept-Graph Reasoning for Interpretable Visual Explanationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026On the Faithfulness of Post-Hoc Concept Bottleneck Models
  2. 2026Learning Concept Bottleneck Models from Mechanistic Explanations
  3. 2026Rethinking Concept Bottleneck Models: From Pitfalls to Solutions
  4. 2023Coarse-to-Fine Concept Bottleneck Models
  5. 2024Concept Bottleneck Models Without Predefined Concepts
  6. 2026Clarity: The Flexibility-Interpretability Trade-Off in Sparsity-aware Concept Bottleneck Models
  7. 2025Partially Shared Concept Bottleneck Models
  8. 2025A Geometric Unification of Concept Learning with Concept Cones