Agustin Martin Picard

Cross-Modal Redundancy and the Geometry of Vision–Language Embeddings

An Adaptive Orthogonal Convolution Scheme for Efficient and Flexible CNN Architectures