mlsanityData to 7 Oct 2026

Representational Difference Explanations

arXiv 2505.23917 ↗First posted 29 May 2025cs.CV cs.AI cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

We propose a method for discovering and visualizing the differences between two learned representations, enabling more direct and interpretable model comparisons. We validate our method, which we call Representational Differences Explanations (RDX), by using it to compare models with known conceptual differences and demonstrate that it recovers meaningful distinctions where existing explainable AI (XAI) techniques fail. Applied to state-of-the-art models on challenging subsets of the ImageNet and iNaturalist datasets, RDX reveals both insightful representational differences and subtle patterns in the data. Although comparison is a cornerstone of scientific analysis, current tools in machine learning, namely post hoc XAI methods, struggle to support model comparison effectively. Our work addresses this gap by introducing an effective and explainable tool for contrasting model representations.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 34 of its 70 references that are in the map. First posted on arXiv 29 May 2025.

Builds on

34 of 70 references.

References · 34

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  3. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,764
  4. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
  5. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  6. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  7. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  8. 2019Similarity of Neural Network Representations Revisitedcited by 1,695
  9. 2023Sparse Autoencoders Find Highly Interpretable Features in Language Modelscited by 1,359
  10. 2017Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)cited by 1,309
  11. 2021Do Vision Transformers See Like Convolutional Neural Networks?cited by 671
  12. 2024Scaling and evaluating sparse autoencoderscited by 530
22 more
  1. 2020What is being transferred in transfer learning?cited by 365
  2. 2024The Platonic Representation Hypothesiscited by 362
  3. 2022Post-hoc Concept Bottleneck Modelscited by 289
  4. 2024Jumping Ahead: Improving Reconstruction Fidelity with JumpReLU Sparse Autoencoderscited by 283
  5. 2020Do Wide and Deep Networks Learn the Same Things? Uncovering How Neural Network Representations Vary with Width and Depthcited by 241
  6. 2015Convergent Learning: Do different neural networks learn the same representations?cited by 215
  7. 2013k-Sparse Autoencoderscited by 209
  8. 2022CRAFT: Concept Recursive Activation FacTorization for Explainabilitycited by 177
  9. 2022From Attribution Maps to Human-Understandable Explanations through Concept Relevance Propagationcited by 141
  10. 2020Invertible Concept-based Explanations for CNN Models with Non-negative Concept Activation Vectorscited by 119
  11. 2022Revealing the Dark Secrets of Masked Image Modelingcited by 117
  12. 2023A Holistic Approach to Unifying Automatic Concept Extraction and Concept Importance Estimationcited by 108
  13. 2021What I Cannot Predict, I Do Not Understand: A Human-Centered Evaluation Framework for Explainability Methodscited by 98
  14. 2025Archetypal SAE: Adaptive and Stable Dictionary Learning for Concept Extraction in Large Vision Modelscited by 67
  15. 2025Universal Sparse Autoencoders: Interpretable Cross-Model Concept Alignmentcited by 62
  16. 2025Mechanistic understanding and validation of large AI models with SemanticLenscited by 29
  17. 2025Interpretable and Testable Vision Features via Sparse Autoencoderscited by 24
  18. 2021Efficient Diversity-Driven Ensemble for Deep Neural Networkscited by 11
  19. 2024Beyond Scalars: Concept-Based Alignment Analysis in Vision Transformerscited by 11
  20. 2025Representational Similarity via Interpretable Visual Conceptscited by 5
  21. 2022DiSC: Differential Spectral Clustering of Featurescited by 4
  22. 2023Training-Free Uncertainty Estimation for Embedding Modelscited by 3

Built on by

Not cited yet.