QEDG improves data-free hard-label model stealing by generating boundary-hugging synthetic samples and reusing a single query to create multiple augmented training examples, beating prior methods with fewer queries.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Exploring Query Efficient Data Generation towards Data-free Model Stealing in Hard Label Setting
QEDG improves data-free hard-label model stealing by generating boundary-hugging synthetic samples and reusing a single query to create multiple augmented training examples, beating prior methods with fewer queries.