Stefano Soatto

Critical Learning Periods Emerge Even in Deep Linear Networks

Tangent Transformers for Composition,Privacy and Removal

Sub-token ViT Embedding via Stochastic Resonance Transformers

Divided Attention: Unsupervised Multiple-object Discovery and Segmentation with Interpretable Contextually Separated Slots

On the Viability of Monocular Depth Pre-training for Semantic Segmentation

AugUndo: Scaling Up Augmentations for Unsupervised Depth Completion

Taming AI Bots: Controllability of Neural States in Large Language Models