mlsanityData to 7 Oct 2026

CHOQOLATE: Organizing Concept Bottleneck Latent Spaces with Choquet Integrals

arXiv 2609.37786 ↗First posted 29 Sep 2026cs.CVcited by 0

On the map: Multimodal LLMs / vision-language models → VLM benchmarks and visual reasoning also fits: Chart understanding and anomaly detection

Abstract (arXiv)

Concept Bottleneck Models (CBMs) built on vision-language models such as CLIP represent a latent space as human-understandable concepts. These representations are unfaithful: related concepts are entangled, so individual scores do not reflect their intended meaning. We propose CHOQOLATE, an interpretable-by-design layer based on 2-additive Choquet integrals, which merges correlated concepts into compact nodes. Across four datasets, CHOQOLATE achieves a favorable accuracy-interpretability trade-off, with weight-sparse and semantically coherent nodes. A closed-form gradient derivation, backed by experiments, explains why Choquet layers drive this organization without explicit supervision. Choquet weights also map directly to Shapley values, which enables test-time intervention. On standard bias-mitigation benchmarks, suppressing spurious concepts after training performs on par with methods that require group annotations or retraining, while needing neither.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in VLM benchmarks and visual reasoning, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 29 of its 53 references that are in the map. First posted on arXiv 29 Sep 2026.

Builds on

29 of 53 references.

References · 29

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  3. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  4. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,040
  5. 2014Deep Learning Face Attributes in the Wildcited by 3,679
  6. 2019Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalizationcited by 1,232
  7. 2020Concept Bottleneck Modelscited by 1,022
  8. 2018This Looks Like That: Deep Learning for Interpretable Image Recognitioncited by 731
  9. 2022When and why vision-language models behave like bags-of-words, and what to do about it?cited by 603
  10. 2022Last Layer Re-Training is Sufficient for Robustness to Spurious Correlationscited by 379
  11. 2024The Platonic Representation Hypothesiscited by 362
  12. 2022Language in a Bottle: Language Model Guided Concept Bottlenecks for Interpretable Image Classificationcited by 310
17 more
  1. 2023Debiasing Vision-Language Models via Biased Promptscited by 147
  2. 2025Learning Multi-Level Features with Matryoshka Sparse Autoencoderscited by 136
  3. 2024Interpreting CLIP with Sparse Linear Concept Embeddings (SpLiCE)cited by 116
  4. 2022Does CLIP Bind Concepts? Probing Compositionality in Large Image Modelscited by 101
  5. 2024Discover-then-Name: Task-Agnostic Concept Bottlenecks via Automated Concept Discoverycited by 91
  6. 2016Places: An Image Database for Deep Scene Understandingcited by 76
  7. 2023Identifying Interpretable Subspaces in Image Representationscited by 32
  8. 2023Coarse-to-Fine Concept Bottleneck Modelscited by 31
  9. 2025If Concept Bottlenecks are the Question, are Foundation Models the Answer?cited by 15
  10. 2023CLIP-QDA: An Explainable Concept Bottleneck Modelcited by 13
  11. 2022A k-additive Choquet integral-based approach to approximate the SHAP values for local interpretability in machine learningcited by 8
  12. 2024Benchmarking XAI Explanations with Human-Aligned Evaluationscited by 8
  13. 2019Enabling Explainable Fusion in Deep Learning with Fuzzy Integral Neural Networkscited by 7
  14. 2023Text Descriptions are Compressive and Invariant Representations for Visual Learningcited by 7
  15. 2025Enhancing Concept Localization in CLIP-based Concept Bottleneck Modelscited by 6
  16. 2025Representation-Level Counterfactual Calibration for Debiased Zero-Shot Recognitioncited by 2
  17. 2026Hoeffding Concept Bottleneck Models with Applications to Overhead Imagescited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Rethinking Concept Bottleneck Models: From Pitfalls to Solutions
  2. 2025Partially Shared Concept Bottleneck Models
  3. 2026On the Faithfulness of Post-Hoc Concept Bottleneck Models
  4. 2026Learning Concept Bottleneck Models from Mechanistic Explanations
  5. 2025Enhancing Interpretable Image Classification Through LLM Agents and Conditional Concept Bottleneck Models
  6. 2024VLG-CBM: Training Concept Bottleneck Models with Vision-Language Guidance
  7. 2024Concept Bottleneck Models Without Predefined Concepts
  8. 2026CBX-Bench: A Human-Aligned MLLM Council for Benchmarking Concept Bottleneck Model Explanations