CVPR 2026 Open Access Repository

How Do Vision Models Work?

I Walk the Line: Examining the Role of Gestalt Continuity in Object Binding for Vision Transformers: Alexa R. Tartaglini,

Michael A Lepori; [pdf] [supp] [arXiv]
[bibtex]
@InProceedings{Tartaglini_2026_CVPR, author = {Tartaglini, Alexa R. and A Lepori, Michael}, title = {I Walk the Line: Examining the Role of Gestalt Continuity in Object Binding for Vision Transformers}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2026}, pages = {4183-4187} }
Differences in Detection: Explainability Where it Matters: Johannes Theodoridis,

Johannes Maucher,

Andreas Schilling; [pdf]
[bibtex]
@InProceedings{Theodoridis_2026_CVPR, author = {Theodoridis, Johannes and Maucher, Johannes and Schilling, Andreas}, title = {Differences in Detection: Explainability Where it Matters}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2026}, pages = {4188-4192} }
LogitDynamics: Reliable ViT Error Detection from Layerwise Logit Trajectories: Ido Beigelman,

Moti Freiman; [pdf] [arXiv]
[bibtex]
@InProceedings{Beigelman_2026_CVPR, author = {Beigelman, Ido and Freiman, Moti}, title = {LogitDynamics: Reliable ViT Error Detection from Layerwise Logit Trajectories}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2026}, pages = {4132-4138} }
Toward Faithful Segmentation Attribution via Benchmarking and Dual-Evidence Fusion: Abu Noman Md Sakib,

OFM Riaz Rahman Aranya,

Kevin Desai,

Zijie Zhang; [pdf] [arXiv]
[bibtex]
@InProceedings{Sakib_2026_CVPR, author = {Sakib, Abu Noman Md and Aranya, OFM Riaz Rahman and Desai, Kevin and Zhang, Zijie}, title = {Toward Faithful Segmentation Attribution via Benchmarking and Dual-Evidence Fusion}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2026}, pages = {4171-4177} }
When Geometry Reverses Topological Conclusions: Evaluating Persistent Homology in Sparse Autoencoders: Teresa Zhang; [pdf]
[bibtex]
@InProceedings{Zhang_2026_CVPR, author = {Zhang, Teresa}, title = {When Geometry Reverses Topological Conclusions: Evaluating Persistent Homology in Sparse Autoencoders}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2026}, pages = {4193-4199} }
Concept Spaces in the Residual Stream of Diffusion Transformers: Riyasat Ohib,

Meera Hahn,

Mani Malek; [pdf] [supp]
[bibtex]
@InProceedings{Ohib_2026_CVPR, author = {Ohib, Riyasat and Hahn, Meera and Malek, Mani}, title = {Concept Spaces in the Residual Stream of Diffusion Transformers}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2026}, pages = {4158-4163} }
Simple Localized Counterfactuals for Visual Explanation: David Carlyn,

Jianyang Gu,

Wei-Lun Chao; [pdf] [supp]
[bibtex]
@InProceedings{Carlyn_2026_CVPR, author = {Carlyn, David and Gu, Jianyang and Chao, Wei-Lun}, title = {Simple Localized Counterfactuals for Visual Explanation}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2026}, pages = {4139-4144} }
Why CNN Features Are not Gaussian: A Statistical Anatomy of Deep Representations: David Chapman,

Parniyan Farvardin; [pdf] [supp] [arXiv]
[bibtex]
@InProceedings{Chapman_2026_CVPR, author = {Chapman, David and Farvardin, Parniyan}, title = {Why CNN Features Are not Gaussian: A Statistical Anatomy of Deep Representations}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2026}, pages = {4145-4150} }
MIMIC: Multimodal Inversion for Model Interpretation and Conceptualization: Animesh Jain,

Alexandros Stergiou; [pdf] [arXiv]
[bibtex]
@InProceedings{Jain_2026_CVPR, author = {Jain, Animesh and Stergiou, Alexandros}, title = {MIMIC: Multimodal Inversion for Model Interpretation and Conceptualization}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2026}, pages = {4151-4157} }
Faithful Attribution in Vision Transformers via Feature-Gradient Gating: Julius Šula,

Thomas Lukasiewicz,

Bayar Menzat; [pdf] [supp]
[bibtex]
@InProceedings{Sula_2026_CVPR, author = {\v{S}ula, Julius and Lukasiewicz, Thomas and Menzat, Bayar}, title = {Faithful Attribution in Vision Transformers via Feature-Gradient Gating}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2026}, pages = {4178-4182} }
Zero-Ablation Overstates Register Content Dependence in DINO Vision Transformers: Felipe Parodi,

Jordan K Matelsky,

Melanie Segado; [pdf] [supp] [arXiv]
[bibtex]
@InProceedings{Parodi_2026_CVPR, author = {Parodi, Felipe and Matelsky, Jordan K and Segado, Melanie}, title = {Zero-Ablation Overstates Register Content Dependence in DINO Vision Transformers}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2026}, pages = {4164-4170} }