A boundary-targeted MIA strategy recovers 19% of distress-flagged conversations from a safety classifier at 5% false-positive rate, 3.5 times better than prior methods.
Yeom Samuel, Giacomelli Irene, Fredrikson Matt, Jha Somesh
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
-
Boundary-targeted Membership Inference Attacks on Safety Classifiers
A boundary-targeted MIA strategy recovers 19% of distress-flagged conversations from a safety classifier at 5% false-positive rate, 3.5 times better than prior methods.