ROSU derives a closed-form retain-neutral perturbation for min-max unlearning that bounds retain damage via curvature and improves performance when gradients are aligned.
Large language model unlearning.Advances in Neural Infor- mation Processing Systems, 37:105425–105475
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
-
Retain-Neutral Surrogates for Min-Max Unlearning
ROSU derives a closed-form retain-neutral perturbation for min-max unlearning that bounds retain damage via curvature and improves performance when gradients are aligned.