3 papers
cs.CV2025
Beyond [cls]: Exploring the true potential of Masked Image Modeling representations
Marcin PrzewiÄźlikowski, Randall Balestriero, Wojciech JasiÅski +2
Masked Image Modeling (MIM) has emerged as a promising approach for Self-Supervised Learning (SSL) of visual representations. However, the out-of-the-box performance of MIMs is typ…
cs.LG2025
A deep cut into Split Federated Self-supervised Learning
Marcin PrzewiÄźlikowski, Marcin Osial, Bartosz ZieliÅski +1
Collaborative self-supervised learning has recently become feasible in highly distributed environments by dividing the network layers between client devices and a central server. H…
cs.CV2024
Augmentation-aware Self-supervised Learning with Conditioned Projector
Marcin PrzewiÄźlikowski, Mateusz Pyla, Bartosz ZieliÅski +3
Self-supervised learning (SSL) is a powerful technique for learning from unlabeled data. By learning to remain invariant to applied data augmentations, methods such as SimCLR and M…