mlsanityData to 7 Oct 2026

Mixture of Concept Bottleneck Experts

arXiv 2602.02886 ↗First posted 2 Feb 2026cs.LG cs.AIcited by 2

Not on the map. Why?

Abstract (arXiv)

Concept Bottleneck Models (CBMs) promote interpretability by grounding predictions in human-understandable concepts. However, existing CBMs typically constrain their task predictor to a single expression whose functional form is set a priori, limiting both predictive accuracy and adaptability to diverse user needs. We propose Mixture of Concept Bottleneck Experts (M-CBE), a framework that generalizes existing CBMs along two dimensions: the number of expressions, referred to as experts, employed by the task predictor to map concepts to the task, and the functional form each expression takes, thus exposing an underexplored region of this design space. We investigate this region by instantiating two novel models: Linear M-CBE, which learns a finite set of linear expressions, and Symbolic M-CBE, which leverages symbolic regression to discover expert functions from data subject to user-specified operator vocabularies. Empirical evaluation demonstrates that varying the number of expressions and their functional form provides a robust framework for navigating the accuracy-interpretability trade-off.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 36 of its 60 references that are in the map. First posted on arXiv 2 Feb 2026.

Builds on

36 of 60 references.

References · 36

  1. 2017Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layercited by 3,413
  2. 2020GShard: Scaling Giant Models with Conditional Computation and Automatic Shardingcited by 1,601
  3. 2017Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)cited by 1,309
  4. 2017Explanation in Artificial Intelligence: Insights from the Social Sciencescited by 1,201
  5. 2020Concept Bottleneck Modelscited by 1,022
  6. 2016The Mythos of Model Interpretabilitycited by 716
  7. 2021Interpretable Machine Learning: Fundamental Principles and 10 Grand Challengescited by 364
  8. 2020Discovering Symbolic Models from Deep Learning with Inductive Biasescited by 335
  9. 2023Label-Free Concept Bottleneck Modelscited by 282
  10. 2017Zero-Shot Learning -- The Good, the Bad and the Uglycited by 254
  11. 2021Neural Symbolic Regression that Scalescited by 185
  12. 2021Efficient Large Scale Language Modeling with Mixtures of Expertscited by 175
24 more
  1. 2018Learning Equations for Extrapolation and Controlcited by 151
  2. 2021Promises and Pitfalls of Black-Box Concept Learning Modelscited by 128
  3. 2016Extrapolation and learning equationscited by 106
  4. 2023Concept-based Explainable Artificial Intelligence: A Surveycited by 97
  5. 2022GlanceNets: Interpretabile, Leak-proof Concept-based Modelscited by 73
  6. 2018How do Humans Understand Explanations from Machine Learning Systems? An Evaluation of the Human-Interpretability of Explanationcited by 71
  7. 2022"Help Me Help the AI": Understanding How Explainability Can Support Human-AI Interactioncited by 63
  8. 2022Concept Embedding Models: Beyond the Accuracy-Explainability Trade-Offcited by 61
  9. 2023Interpretable Neural-Symbolic Concept Reasoningcited by 56
  10. 2024Stochastic Concept Bottleneck Modelscited by 49
  11. 2020CausalVAE: Structured Causal Disentanglement in Variational Autoencodercited by 47
  12. 2021Logic Explained Networkscited by 45
  13. 2024Causal Concept Graph Models: Beyond Causal Opacity in Deep Learningcited by 28
  14. 2017Fine-grained Visual-textual Representation Learningcited by 27
  15. 2021Preferential Mixture-of-Experts: Interpretable Models that Rely on Human Expertise as much as Possiblecited by 19
  16. 2023Controllable Neural Symbolic Regressioncited by 19
  17. 2022Interpretable Mixture of Expertscited by 17
  18. 2024Interpretable Concept-Based Memory Reasoningcited by 17
  19. 2025Causally Reliable Concept Bottleneck Modelscited by 17
  20. 2024Concept Bottleneck Language Models For protein designcited by 15
  21. 2025If Concept Bottlenecks are the Question, are Foundation Models the Answer?cited by 15
  22. 2025Avoiding Leakage Poisoning: Concept Interventions Under Distribution Shiftscited by 9
  23. 2025Mathematical Foundation of Interpretable Equivariant Surrogate Modelscited by 5
  24. 2024Linearly-Interpretable Concept Embedding Models for Text Analysiscited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026ConceptM3oE: Concept-Guided Multimodal Mixture of Experts for Interpretable Computational Pathologycited by 0
  2. 2026Graphical Design of Interpretable Architecturescited by 0