mlsanityData to 7 Oct 2026

Selective Concept Models: Permitting Stakeholder Customisation at Test-Time

arXiv 2306.08424 ↗First posted 14 Jun 2023cs.HC cs.AI cs.LGcited by 6

Not on the map. Why?

Abstract (arXiv)

Concept-based models perform prediction using a set of concepts that are interpretable to stakeholders. However, such models often involve a fixed, large number of concepts, which may place a substantial cognitive load on stakeholders. We propose Selective COncept Models (SCOMs) which make predictions using only a subset of concepts and can be customised by stakeholders at test-time according to their preferences. We show that SCOMs only require a fraction of the total concepts to achieve optimal accuracy on multiple real-world datasets. Further, we collect and release a new dataset, CUB-Sel, consisting of human concept set selections for 900 bird images from the popular CUB dataset. Using CUB-Sel, we show that humans have unique individual preferences for the choice of concepts they prefer to reason about, and struggle to identify the most theoretically informative concepts. The customisation and concept selection provided by SCOM improves the efficiency of interpretation and intervention for stakeholders.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 21 of its 46 references that are in the map. First posted on arXiv 14 Jun 2023.

Builds on

21 of 46 references.

References · 21

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Deep Learning Face Attributes in the Wildcited by 3,679
  3. 2017Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)cited by 1,309
  4. 2020Concept Bottleneck Modelscited by 1,022
  5. 2018Towards Robust Interpretability with Self-Explaining Neural Networkscited by 505
  6. 2017The Effectiveness of Data Augmentation in Image Classification using Deep Learningcited by 368
  7. 2022Post-hoc Concept Bottleneck Modelscited by 289
  8. 2018Learning to Explain: An Information-Theoretic Perspective on Model Interpretationcited by 248
  9. 2019On Completeness-aware Concept-Based Explanations in Deep Neural Networkscited by 239
  10. 2019Human uncertainty makes classification more robustcited by 231
  11. 2022Interactive Concept Bottleneck Modelscited by 79
  12. 2022GlanceNets: Interpretabile, Leak-proof Concept-based Modelscited by 73
9 more
  1. 2022"Help Me Help the AI": Understanding How Explainability Can Support Human-AI Interactioncited by 63
  2. 2023A Closer Look at the Intervention Procedure of Concept Bottleneck Modelscited by 63
  3. 2022Concept Embedding Models: Beyond the Accuracy-Explainability Trade-Offcited by 61
  4. 2022Concept Activation Regions: A Generalized Framework For Concept-Based Explanationscited by 57
  5. 2021Meaningfully Debugging Model Mistakes using Conceptual Counterfactual Explanationscited by 53
  6. 2022Eliciting and Learning with Soft Labels from Every Annotatorcited by 43
  7. 2023Human Uncertainty in Concept-Based AI Systemscited by 39
  8. 2017Estimating Mixture Entropy with Pairwise Distancescited by 34
  9. 2022Concept Gradient: Concept-based Interpretation Without Linear Assumptioncited by 20

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 2; 2024: 1; 2025: 2; 2026: 1.

6 papers.

  1. 2023Learning to Receive Help: Intervention-Aware Concept Embedding Modelscited by 41
  2. 2026Actionable Interpretability Must Be Defined in Terms of Symmetriescited by 4
  3. 2025Foundations of Interpretable Modelscited by 3
  4. 2024Beyond Thumbs Up/Down: Untangling Challenges of Fine-Grained Feedback for Text-to-Image Generationcited by 2
  5. 2023Estimation of Concept Explanations Should be Uncertainty Awarecited by 1
  6. 2025The Impact of Concept Explanations and Interventions on Human-Machine Collaborationcited by 0