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

Hongyan Chang¹, Ali Shahin Shamsabadi², Kleomenis Katevas², Hamed Haddadi^2,3, Reza Shokri⁴

¹Mohamed bin Zayed University of Artificial Intelligence, ²Brave, ³Imperial College London ⁴National University of Singapore

Paper Code

We propose CAMIA (Context-Aware Membership Inference Attack), a simple and effective method to infer if a large language model was pretrained on the given text.

BibTeX

@article{chang2024context,
  title={Context-aware membership inference attacks against pre-trained large language models},
  author={Chang, Hongyan and Shamsabadi, Ali Shahin and Katevas, Kleomenis and Haddadi, Hamed and Shokri, Reza},
  booktitle={Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP)},
  year={2025}
}