Self-supervised ReLU networks form substantially fewer linear regions than supervised models for comparable accuracy, with contrastive methods rapidly expanding regions and self-distillation consolidating them, enabling early geometric detection of representation collapse.
Emerg- ing properties in self-supervised vision transformers
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
verdicts
UNVERDICTED 5representative citing papers
Pretrained vision transformers use specific attention heads sensitive to Gestalt continuity for object binding, shown via probes on synthetic datasets and ablation experiments.
Chorus pretrains a shared 3D Gaussian scene encoder via multi-teacher distillation to capture holistic features from high-level semantics to fine-grained structure, with strong transfer on segmentation and point-cloud tasks using far fewer scenes.
A multi-teacher collaborative framework with reliability assessment for forward-looking sonar semantic segmentation reports 5.08% mIoU gain on FLSMD dataset using only 2% labeled data.
SAM 3 can be applied training-free to remote sensing open-vocabulary segmentation and change detection by fusing its semantic and instance heads and filtering with presence scores.
citing papers explorer
-
Complexity of Linear Regions in Self-supervised Deep ReLU Networks
Self-supervised ReLU networks form substantially fewer linear regions than supervised models for comparable accuracy, with contrastive methods rapidly expanding regions and self-distillation consolidating them, enabling early geometric detection of representation collapse.
-
I Walk the Line: Examining the Role of Gestalt Continuity in Object Binding for Vision Transformers
Pretrained vision transformers use specific attention heads sensitive to Gestalt continuity for object binding, shown via probes on synthetic datasets and ablation experiments.
-
Chorus: Multi-Teacher Pretraining for Holistic 3D Gaussian Scene Encoding
Chorus pretrains a shared 3D Gaussian scene encoder via multi-teacher distillation to capture holistic features from high-level semantics to fine-grained structure, with strong transfer on segmentation and point-cloud tasks using far fewer scenes.
-
CTFS : Collaborative Teacher Framework for Forward-Looking Sonar Image Semantic Segmentation with Extremely Limited Labels
A multi-teacher collaborative framework with reliability assessment for forward-looking sonar semantic segmentation reports 5.08% mIoU gain on FLSMD dataset using only 2% labeled data.
-
SegEarth-OV3: Exploring SAM 3 for Open-Vocabulary Semantic Segmentation in Remote Sensing Images
SAM 3 can be applied training-free to remote sensing open-vocabulary segmentation and change detection by fusing its semantic and instance heads and filtering with presence scores.