Standard model inversion evaluation counts many adversarial false positives as successes; MLLM-based evaluation reveals consistently high false-positive rates across 27 attack setups.
Adversarial attack type i: Cheat classifiers by significant changes.IEEE transactions on pattern analysis and machine intelligence, 43(3):1100–1109
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
-
Revisiting Model Inversion Evaluation: From Misleading Standards to Reliable Privacy Assessment
Standard model inversion evaluation counts many adversarial false positives as successes; MLLM-based evaluation reveals consistently high false-positive rates across 27 attack setups.