Pith. sign in

Exploring simple siamese rep- resentation learning

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

fields

cs.CV 1

years

2024 1

verdicts

CONDITIONAL 1

representative citing papers

Multi-Token Enhancing for Vision Representation Learning

cs.CV · 2024-11-24 · conditional · novelty 6.0

Training a vision transformer with multiple auxiliary tokens and distilling them into a single global token improves self-supervised representation quality with no additional inference cost.

citing papers explorer

Showing 1 of 1 citing paper.

  • Multi-Token Enhancing for Vision Representation Learning cs.CV · 2024-11-24 · conditional · none · ref 8

    Training a vision transformer with multiple auxiliary tokens and distilling them into a single global token improves self-supervised representation quality with no additional inference cost.