Attention-weighted negative sampling with LSH sparsification improves bias removal in contrastive representation learning while approximately preserving accuracy.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
An Attention-based Framework for Fair Contrastive Learning
Attention-weighted negative sampling with LSH sparsification improves bias removal in contrastive representation learning while approximately preserving accuracy.