mlsanityData to 14 Oct 2026

Interpreting and Controlling Vision Foundation Models via Text Explanations

arXiv 2310.10591 ↗First posted 16 Oct 2023cs.CVcited by 11

Not on the map. Why?

Abstract (arXiv)

Large-scale pre-trained vision foundation models, such as CLIP, have become de facto backbones for various vision tasks. However, due to their black-box nature, understanding the underlying rules behind these models' predictions and controlling model behaviors have remained open challenges. We present a framework for interpreting vision transformer's latent tokens with natural language. Given a latent token, our framework retains its semantic information to the final layer using transformer's local operations and retrieves the closest text for explanation. Our approach enables understanding of model visual reasoning procedure without needing additional model training or data collection. Based on the obtained interpretations, our framework allows for model editing that controls model reasoning behaviors and improves model robustness against biases and spurious correlations.

Key facts. Cited by 11 papers in the map (data through 14 Oct 2026). Builds on 51 of its 62 references that are in the map. First posted on arXiv 16 Oct 2023.

Builds on

51 of 62 references.

References · 51

  1. 2017Attention Is All You Needcited by 60,361
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  3. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  4. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  5. 2023Segment Anythingcited by 8,234
  6. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
  7. 2017A Unified Approach to Interpreting Model Predictionscited by 6,555
  8. 2023BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Modelscited by 6,137
  9. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,070
  10. 2014Neural Machine Translation by Jointly Learning to Align and Translatecited by 4,548
  11. 2021Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervisioncited by 3,864
  12. 2014Deep Learning Face Attributes in the Wildcited by 3,693
38 more
  1. 2023MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Modelscited by 2,711
  2. 2022Locating and Editing Factual Associations in GPTcited by 2,691
  3. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,371
  4. 2013Visualizing and Understanding Convolutional Networkscited by 2,290
  5. 2020Shortcut Learning in Deep Neural Networkscited by 1,969
  6. 2022Visual Prompt Tuningcited by 1,681
  7. 2017Learning Important Features Through Propagating Activation Differencescited by 1,458
  8. 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,340
  9. 2017Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)cited by 1,319
  10. 2023ImageBind: One Embedding Space To Bind Them Allcited by 1,175
  11. 2017SmoothGrad: removing noise by adding noisecited by 1,109
  12. 2020Concept Bottleneck Modelscited by 1,032
  13. 2021Open-vocabulary Object Detection via Vision and Language Knowledge Distillationcited by 906
  14. 2020Quantifying Attention Flow in Transformerscited by 775
  15. 2017Network Dissection: Quantifying Interpretability of Deep Visual Representationscited by 708
  16. 2021FLAVA: A Foundational Language And Vision Alignment Modelcited by 649
  17. 2021VideoCLIP: Contrastive Pre-training for Zero-shot Video-Text Understandingcited by 602
  18. 2018Do Better ImageNet Models Transfer Better?cited by 565
  19. 2022Socratic Models: Composing Zero-Shot Multimodal Reasoning with Languagecited by 513
  20. 2014Understanding Deep Image Representations by Inverting Themcited by 403
  21. 2019Provably Robust Deep Learning via Adversarially Trained Smoothed Classifierscited by 330
  22. 2022GLIPv2: Unifying Localization and Vision-Language Understandingcited by 275
  23. 2018Textual Explanations for Self-Driving Vehiclescited by 226
  24. 2017Exploring Models and Data for Remote Sensing Image Caption Generationcited by 197
  25. 2023Interpreting CLIP's Image Representation via Text-Based Decompositioncited by 196
  26. 2016Synthesizing the preferred inputs for neurons in neural networks via deep generator networkscited by 181
  27. 2016Generating Visual Explanationscited by 158
  28. 2022Natural Language Descriptions of Deep Visual Featurescited by 137
  29. 2022Understanding Zero-Shot Adversarial Robustness for Large-Scale Modelscited by 117
  30. 2020Rewriting a Deep Generative Modelcited by 107
  31. 2021Learning to Predict Visual Attributes in the Wildcited by 96
  32. 2021Editing a classifier by rewriting its prediction rulescited by 65
  33. 2018Not Using the Car to See the Sidewalk: Quantifying and Controlling the Effects of Context in Classification and Segmentationcited by 49
  34. 2020Generative Interventions for Causal Learningcited by 45
  35. 2020Debiasing Skin Lesion Datasets and Models? Not So Fastcited by 43
  36. 2019Certifiably Robust Interpretation in Deep Learningcited by 35
  37. 2021Discrete Representations Strengthen Vision Transformer Robustnesscited by 35
  38. 2022Fine-tuning Image Transformers using Learnable Memorycited by 34

50 most cited of 51 All 51 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 3; 2025: 5; 2026: 3.

11 papers.

  1. 2024Interpreting and Editing Vision-Language Representations to Mitigate Hallucinationscited by 104
  2. 2024Discover-then-Name: Task-Agnostic Concept Bottlenecks via Automated Concept Discoverycited by 91
  3. 2025How Visual Representations Map to Language Feature Space in Multimodal LLMscited by 25
  4. 2025To Sink or Not to Sink: Visual Information Pathways in Large Vision-Language Modelscited by 20
  5. 2024Decomposing and Editing Predictions by Modeling Model Computationcited by 19
  6. 2025Too Late to Recall: Explaining the Two-Hop Problem in Multimodal Knowledge Retrievalcited by 7
  7. 2025Knowledge Transfer from Interaction Learningcited by 2
  8. 2026Towards Understanding Multimodal Fine-Tuning: Spatial Featurescited by 1
  9. 2025TRANSPORTER: Transferring Visual Semantics from VLM Manifoldscited by 0
  10. 2026Multimodal Model Diffing for Feature Discovery and Controlcited by 0
  11. 2026Using OCR Heads to Verbalize Image Semanticscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023From Wrong To Right: A Recursive Approach Towards Vision-Language Explanation
  2. 2026Rethinking Latent Visual Reasoning: Grounding Latent Reasoning in Visual Evidence
  3. 2026Decompose, Look, and Reason: Reinforced Latent Reasoning for VLMs
  4. 2026Right Predictions, Misleading Explanations: On the Vulnerability of Vision-Language Model Explanations
  5. 2026Retrieve, Integrate, and Synthesize: Spatial-Semantic Grounded Latent Visual Reasoning
  6. 2024GIFT: A Framework Towards Global Interpretable Faithful Textual Explanations of Vision Classifiers
  7. 2026Do Vision-Language Models Truly Perform Vision Reasoning? A Rigorous Study of the Modality Gap
  8. 2026LanteRn: Latent Visual Structured Reasoning