minus-squaremileseverett@alien.topBtoMachine Learning@academy.garden•[R] ConvNets Match Vision Transformers at ScalelinkfedilinkEnglisharrow-up1·11 months agoMasked Image Modelling objectives are just harder with CNNs compared to ViTs linkfedilink
Masked Image Modelling objectives are just harder with CNNs compared to ViTs