Foundation models yield less human-interpretable features than supervised vision transformers, with interpretability tied to activation locality and coarse semantic alignment rather than task performance.
Evaluating general purpose vision foundation models for medical image analysis: An experimental study of dinov2 on radiology benchmarks
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
years
2026 7representative citing papers
Proposes LHE, SRB, and AFL components in a semantics-first latent framework that yields better 3D MRI reconstruction and cross-contrast synthesis on two public datasets.
DALE-CT, a 2D LeJEPA model with depth-aware dual supervision, reaches 0.833 Macro AUROC on multi-abnormality detection in CT and approaches 3D SOTA performance using less data and no textual supervision.
VoxCor creates reusable volumetric features from frozen 2D ViT models by combining triplanar inference with a closed-form weighted partial least squares projection, enabling direct voxel correspondence across modalities without training or registration.
Natural-domain foundation models provide competitive and more robust priors than task-specific models for accelerated cardiac MRI reconstruction in cross-domain settings.
A vision transformer classifier trained on simulated and real Euclid data recovers all known strong lenses in test sets and finds 8 Grade A plus 26 Grade B new candidates in the Q1 data.
DINOCell achieves a SEG score of 0.784 on LIVECell by self-supervised domain adaptation of DINOv2, improving 10.42% over SAM-based models and showing strong zero-shot transfer.
citing papers explorer
-
Capability $\neq$ Interpretability: Human Interpretability of Vision Foundation Models
Foundation models yield less human-interpretable features than supervised vision transformers, with interpretability tied to activation locality and coarse semantic alignment rather than task performance.
-
Recover Semantics First, Generate Better: Improved Latent Modeling for 3D MRI Reconstruction and Cross-Contrast Synthesis
Proposes LHE, SRB, and AFL components in a semantics-first latent framework that yields better 3D MRI reconstruction and cross-contrast synthesis on two public datasets.
-
DALE-CT: Depth-Aware Foundation Models for Computed Tomography
DALE-CT, a 2D LeJEPA model with depth-aware dual supervision, reaches 0.833 Macro AUROC on multi-abnormality detection in CT and approaches 3D SOTA performance using less data and no textual supervision.
-
VoxCor: Training-Free Volumetric Features for Multimodal Voxel Correspondence
VoxCor creates reusable volumetric features from frozen 2D ViT models by combining triplanar inference with a closed-form weighted partial least squares projection, enabling direct voxel correspondence across modalities without training or registration.
-
Are Natural-Domain Foundation Models Effective for Accelerated Cardiac MRI Reconstruction?
Natural-domain foundation models provide competitive and more robust priors than task-specific models for accelerated cardiac MRI reconstruction in cross-domain settings.
-
Euclid Quick Data Release (Q1). AstroVink: A vision transformer approach to find strong gravitational lens systems
A vision transformer classifier trained on simulated and real Euclid data recovers all known strong lenses in test sets and finds 8 Grade A plus 26 Grade B new candidates in the Q1 data.
-
Self-supervised Pretraining of Cell Segmentation Models
DINOCell achieves a SEG score of 0.784 on LIVECell by self-supervised domain adaptation of DINOv2, improving 10.42% over SAM-based models and showing strong zero-shot transfer.