Sep 15, 2026 Where Does a Vision Transformer Keep Its Sense of Geometry? Mar 06, 2026 Autoencoders: From Reconstruction to Representation Learning