mlsanityData to 7 Oct 2026

Concept-Based Explanations in Computer Vision: Where Are We and Where Could We Go?

arXiv 2409.13456 ↗First posted 20 Sep 2024cs.CVcited by 17

Not on the map. Why?

Abstract (arXiv)

Concept-based XAI (C-XAI) approaches to explaining neural vision models are a promising field of research, since explanations that refer to concepts (i.e., semantically meaningful parts in an image) are intuitive to understand and go beyond saliency-based techniques that only reveal relevant regions. Given the remarkable progress in this field in recent years, it is time for the community to take a critical look at the advances and trends. Consequently, this paper reviews C-XAI methods to identify interesting and underexplored areas and proposes future research directions. To this end, we consider three main directions: the choice of concepts to explain, the choice of concept representation, and how we can control concepts. For the latter, we propose techniques and draw inspiration from the field of knowledge representation and learning, showing how this could enrich future C-XAI research.

Key facts. Cited by 17 papers in the map (data through 7 Oct 2026). Builds on 94 of its 132 references that are in the map. First posted on arXiv 20 Sep 2024.

Builds on

94 of 132 references.

References · 94

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  3. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  4. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,415
  5. 2018UMAP: Uniform Manifold Approximation and Projection for Dimension Reductioncited by 2,841
  6. 2016Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotationscited by 2,646
  7. 2017Towards A Rigorous Science of Interpretable Machine Learningcited by 1,400
  8. 2023Sparse Autoencoders Find Highly Interpretable Features in Language Modelscited by 1,359
  9. 2017Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)cited by 1,309
  10. 2016ConceptNet 5.5: An Open Multilingual Graph of General Knowledgecited by 1,065
  11. 2020Concept Bottleneck Modelscited by 1,022
  12. 2023A Comprehensive Survey of Continual Learning: Theory, Method and Applicationcited by 860
38 more
  1. 2018This Looks Like That: Deep Learning for Interpretable Image Recognitioncited by 731
  2. 2017Network Dissection: Quantifying Interpretability of Deep Visual Representationscited by 703
  3. 2018Towards Robust Interpretability with Self-Explaining Neural Networkscited by 505
  4. 2023Unified Concept Editing in Diffusion Modelscited by 377
  5. 2023Explainability for Large Language Models: A Surveycited by 331
  6. 2022Post-hoc Concept Bottleneck Modelscited by 289
  7. 2023Label-Free Concept Bottleneck Modelscited by 282
  8. 2017A Semantic Loss Function for Deep Learning with Symbolic Knowledgecited by 267
  9. 2020Neurosymbolic AI: The 3rd Wavecited by 253
  10. 2020Opportunities and Challenges in Explainable Artificial Intelligence (XAI): A Surveycited by 252
  11. 2019On Completeness-aware Concept-Based Explanations in Deep Neural Networkscited by 239
  12. 2023Evaluating the Ripple Effects of Knowledge Editing in Language Modelscited by 238
  13. 2017Deep Learning for Case-Based Reasoning through Prototypes: A Neural Network that Explains Its Predictionscited by 234
  14. 2020Concept Whitening for Interpretable Image Recognitioncited by 225
  15. 2020Logic Tensor Networkscited by 197
  16. 2022CRAFT: Concept Recursive Activation FacTorization for Explainabilitycited by 177
  17. 2018Visual Interpretability for Deep Learning: a Surveycited by 155
  18. 2021Self-supervised Video Object Segmentation by Motion Groupingcited by 147
  19. 2023Infinite Photorealistic Worlds using Procedural Generationcited by 142
  20. 2022From Attribution Maps to Human-Understandable Explanations through Concept Relevance Propagationcited by 141
  21. 2019Explaining Classifiers with Causal Concept Effect (CaCE)cited by 135
  22. 2024Infinigen Indoors: Photorealistic Indoor Scenes using Procedural Generationcited by 129
  23. 2018Net2Vec: Quantifying and Explaining how Concepts are Encoded by Filters in Deep Neural Networkscited by 128
  24. 2021Promises and Pitfalls of Black-Box Concept Learning Modelscited by 128
  25. 2022Toward Transparent AI: A Survey on Interpreting the Inner Structures of Deep Neural Networkscited by 126
  26. 2021Explainable AI (XAI): A Systematic Meta-Survey of Current Challenges and Future Opportunitiescited by 125
  27. 2020Invertible Concept-based Explanations for CNN Models with Non-negative Concept Activation Vectorscited by 119
  28. 2021Do Concept Bottleneck Models Learn as Intended?cited by 114
  29. 2023A Holistic Approach to Unifying Automatic Concept Extraction and Concept Importance Estimationcited by 108
  30. 2023Concept-based Explainable Artificial Intelligence: A Surveycited by 97
  31. 2017Logic Tensor Networks for Semantic Image Interpretationcited by 92
  32. 2021A Comprehensive Taxonomy for Explainable Artificial Intelligence: A Systematic Survey of Surveys on Methods and Conceptscited by 90
  33. 2020ProtoPShare: Prototype Sharing for Interpretable Image Classification and Similarity Discoverycited by 89
  34. 2022MetaShift: A Dataset of Datasets for Evaluating Contextual Distribution Shifts and Training Conflictscited by 81
  35. 2022Concept Bottleneck Model with Additional Unsupervised Conceptscited by 79
  36. 2022GlanceNets: Interpretabile, Leak-proof Concept-based Modelscited by 73
  37. 2021This Looks Like That... Does it? Shortcomings of Latent Space Prototype Interpretability in Deep Networkscited by 65
  38. 2021Editing a classifier by rewriting its prediction rulescited by 64

50 most cited of 94 All 94 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 9; 2026: 8.

17 papers.

Sort:

By area:

  1. 2025On Background Bias of Post-Hoc Concept Embeddings in Computer Vision DNNscited by 1
  2. 2025Activation Matching for Explanation Generationcited by 1
  3. 2025When concept-based XAI is imprecise: Do people distinguish between generalisations and misrepresentations?cited by 0
  4. 2025On the Performance of Concept Probing: The Influence of the Data (Extended Version)cited by 0
  5. 2025Chat-CBM: Towards Interactive Concept Bottleneck Models with Frozen Large Language Modelscited by 0
  6. 2025ConceptFlow: Hierarchical and Fine-grained Concept-Based Explanation for Convolutional Neural Networkscited by 0
  7. 2025EXP-CAM: Explanation Generation and Circuit Discovery Using Classifier Activation Matchingcited by 0
  8. 2025Attack logics, not outputs: Towards efficient robustification of deep neural networks by falsifying concept-based propertiescited by 0
  9. 2025Towards Safer and Understandable Driver Intention Predictioncited by 0
  10. 2026Ensembles-based Feature Guided Analysiscited by 0
  11. 2026Explaining, Verifying, and Aligning Semantic Hierarchies in Vision-Language Model Embeddingscited by 0
  12. 2026Post-hoc Self-explanation of CNNscited by 0
5 more
  1. 2026Engineering Resource-constrained Software Systems with DNN Components: a Concept-based Pruning Approachcited by 0
  2. 2026Med-CAM: Minimal Evidence for Explaining Medical Decision Makingcited by 0
  3. 2026Bias Leaves a Gradient Trail: Label-Free Bias Identification via Gradient Probes on Concept Decompositionscited by 0
  4. 2026ConceptTree: Bringing Semantic Transparency to Black-Box Decision Making for Robotic Manipulationcited by 0
  5. 2026Identifying Confusion Trends in Concept-based XAI for Multi-Label Classificationcited by 0