Kernel surrogate models with first-order gradient approximation achieve 25% higher correlation to leave-one-out ground truth for task attribution and 40% better downstream data selection than linear surrogates.
Provided that the random projection dimensionksatisfiesk=O logN ϵ2 , the training loss ofcW(S) is bounded away from the minimum training loss for anyS⊆ {1,2
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Efficient Estimation of Kernel Surrogate Models for Task Attribution
Kernel surrogate models with first-order gradient approximation achieve 25% higher correlation to leave-one-out ground truth for task attribution and 40% better downstream data selection than linear surrogates.