Context-Aware Membership Inference Attacks against Pre-trained Large Language Models

Hongyan Chang, Ali Shahin Shamsabadi, Kleomenis Katevas, Hamed Haddadi, Reza Shokri
9/11/2024
cs.CLcs.AIcs.CR

No confident code implementation found for this paper. 8 possible matches can be shown below.

Abstract

Membership Inference Attacks (MIAs) on pre-trained Large Language Models (LLMs) aim at determining if a data point was part of the model's training set. Prior MIAs that are built for classification models fail at LLMs, due to ignoring the generative nature of LLMs across token sequences. In this paper, we present a novel attack on pre-trained LLMs that adapts MIA statistical tests to the perplexity dynamics of subsequences within a data point. Our method significantly outperforms prior approaches, revealing context-dependent memorization patterns in pre-trained LLMs.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. 8 weaker matches are hidden by default — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Cite this paper

@article{chang2024contextaware,
  title  = {Context-Aware Membership Inference Attacks against Pre-trained Large Language Models},
  author = {Hongyan Chang and Ali Shahin Shamsabadi and Kleomenis Katevas and Hamed Haddadi and Reza Shokri},
  year   = {2024},
  eprint = {2409.13745},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2409.13745v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2409.13745.svg)](https://codeofpaper.com/paper/2409.13745)

Discussion