Influence-SSL defines influence as the sensitivity of a sample's representation to augmentation, and shows the resulting scores can identify duplicates, outliers, and fairness-relevant examples in SSL models.
On second- order group influence functions for black-box predictions
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Where Did Your Model Learn That? Label-free Influence for Self-supervised Learning
Influence-SSL defines influence as the sensitivity of a sample's representation to augmentation, and shows the resulting scores can identify duplicates, outliers, and fairness-relevant examples in SSL models.