LoMime: Query-Efficient Membership Inference using Model Extraction in Label-Only Settings
This work addresses privacy threats in machine learning by making membership inference attacks more practical under strict black-box constraints, though it is incremental as it builds on existing label-only MIA approaches.
The paper tackles the problem of high query costs in label-only membership inference attacks by proposing a framework based on model extraction, which reduces the query budget to about 1% of training samples while maintaining accuracy within ±1% of state-of-the-art methods on benchmarks like Purchase, Location, and Texas Hospital.
Membership inference attacks (MIAs) threaten the privacy of machine learning models by revealing whether a specific data point was used during training. Existing MIAs often rely on impractical assumptions such as access to public datasets, shadow models, confidence scores, or training data distribution knowledge and making them vulnerable to defenses like confidence masking and adversarial regularization. Label-only MIAs, even under strict constraints suffer from high query requirements per sample. We propose a cost-effective label-only MIA framework based on transferability and model extraction. By querying the target model M using active sampling, perturbation-based selection, and synthetic data, we extract a functionally similar surrogate S on which membership inference is performed. This shifts query overhead to a one-time extraction phase, eliminating repeated queries to M . Operating under strict black-box constraints, our method matches the performance of state-of-the-art label-only MIAs while significantly reducing query costs. On benchmarks including Purchase, Location, and Texas Hospital, we show that a query budget equivalent to testing $\approx1\%$ of training samples suffices to extract S and achieve membership inference accuracy within $\pm1\%$ of M . We also evaluate the effectiveness of standard defenses proposed for label-only MIAs against our attack.