Consistency models achieve fast one-step generation with SOTA FID of 3.55 on CIFAR-10 and 6.20 on ImageNet 64x64 by directly mapping noise to data, outperforming prior distillation techniques.
Bootstrap your own latent-a new approach to self-supervised learning
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 2roles
background 1polarities
background 1representative citing papers
CurvSSL augments Barlow Twins-style SSL with a curvature alignment loss computed from k-nearest-neighbor cosine scores on the unit hypersphere, yielding competitive linear evaluation accuracy on MNIST and CIFAR-10.
citing papers explorer
-
Consistency Models
Consistency models achieve fast one-step generation with SOTA FID of 3.55 on CIFAR-10 and 6.20 on ImageNet 64x64 by directly mapping noise to data, outperforming prior distillation techniques.
-
Self-Supervised Learning by Curvature Alignment
CurvSSL augments Barlow Twins-style SSL with a curvature alignment loss computed from k-nearest-neighbor cosine scores on the unit hypersphere, yielding competitive linear evaluation accuracy on MNIST and CIFAR-10.