Semantic Membership Inference Attack against Large Language Models
Mozaffari, Hamid, Marathe, Virendra J.
–arXiv.org Artificial Intelligence
Membership Inference Attacks (MIAs) determine whether a specific data point was included in the training set of a target model. In this paper, we introduce the Semantic Membership Inference Attack (SMIA), a novel approach that enhances MIA performance by leveraging the semantic content of inputs and their perturbations. SMIA trains a neural network to analyze the target model's behavior on perturbed inputs, effectively capturing variations in output probability distributions between members and non-members. We conduct comprehensive evaluations on the Pythia and GPT-Neo model families using the Wikipedia dataset. Our results show that SMIA significantly outperforms existing MIAs; for instance, SMIA achieves an AUC-ROC of 67.39% on Pythia-12B, compared to 58.90% by the second-best attack.
arXiv.org Artificial Intelligence
Jun-14-2024
- Country:
- North America > United States > California (0.04)
- Genre:
- Research Report > New Finding (0.87)
- Industry:
- Law (1.00)
- Information Technology > Security & Privacy (1.00)
- Technology: