Soft-weighted unlearning replaces binary data removal with per-sample weights from a convex quadratic program, improving fairness and robustness gains while preserving utility.
A duty to forget, a right to be assured? exposing vulnerabilities in machine unlearning services
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Soft Weighted Machine Unlearning
Soft-weighted unlearning replaces binary data removal with per-sample weights from a convex quadratic program, improving fairness and robustness gains while preserving utility.