Semantic Membership Inference Attack against Large Language Models

2024-06-14 · Hamid Mozaffari, Virendra J. Marathe

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.

📄 PDF Abstract BibTeX arXiv:2406.10218

Code (0)

등록된 구현이 없습니다.

Tasks

Inference AttackMembership Inference Attack

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically

GPT-Neo An implementation of model & data parallel GPT3-like models using the mesh-tensorflow…

Pythia Pythia is a suite of decoder-only autoregressive language models all trained on public data seen in the exact same order and ranging in size from 70M to 12B parameters. The…

Segmentations-Leak: Membership Inference Attacks and Defenses in Semantic Image Segmentation

2019-12-20 · ECCV 2020 8 · Yang He, Shadi Rahimian, Bernt Schiele, Mario Fritz

Today's success of state of the art methods for semantic segmentation is driven by large datasets. Data is considered an important asset that needs to be protected, as the collection and annotation of such datasets comes…

BIG-bench Machine Learning Image Segmentation Segmentation Semantic Segmentation

Bounding Training Data Reconstruction in Private (Deep) Learning

2022-01-28 · Chuan Guo, Brian Karrer, Kamalika Chaudhuri, Laurens van der Maaten

Differential privacy is widely accepted as the de facto method for preventing data leakage in ML, and conventional wisdom suggests that it offers strong protection against privacy attacks. However, existing semantic guar…

Deep Learning

Reconstruction and Membership Inference Attacks against Generative Models

2019-06-07 · Benjamin Hilprecht, Martin Härterich, Daniel Bernau

We present two information leakage attacks that outperform previous work on membership inference against generative models. The first attack allows membership inference without assumptions on the type of the generative m…

Density Estimation Inference Attack Membership Inference Attack

Privacy Analysis of Deep Learning in the Wild: Membership Inference Attacks against Transfer Learning

2020-09-10 · Yang Zou, Zhikun Zhang, Michael Backes, Yang Zhang

While being deployed in many critical applications as core components, machine learning (ML) models are vulnerable to various security and privacy attacks. One major privacy attack in this domain is membership inference,…

BIG-bench Machine Learning Transfer Learning

Semi-Leak: Membership Inference Attacks Against Semi-supervised Learning

2022-07-25 · Xinlei He, Hongbin Liu, Neil Zhenqiang Gong, Yang Zhang

Semi-supervised learning (SSL) leverages both labeled and unlabeled data to train machine learning (ML) models. State-of-the-art SSL methods can achieve comparable performance to supervised learning by leveraging much fe…

Data Augmentation Inference Attack Membership Inference Attack