paperMay 2026Unreviewed

Pretraining Data Exposure in Large Language Models: A Survey of Membership Inference, Data Contamination, and Security Implications

Ziyi Tong, Feifei Sun, Le Minh Nguyen

Abstract

Large Language Models (LLMs) have become the predominant paradigm in NLP, advancing both research and industry. As model sizes and pretraining data grow, concerns about Pretraining Data Exposure (PDE) increase due to the scale and opacity of training datasets. PDE refers to determining whether specific data appeared in an LLM's pretraining corpus. It is critical for ensuring evaluation integrity and protecting privacy, intersecting two key areas: data contamination and membership inference. Thou

Framework mappings

OWASP Top 10 for LLM Applications

LLM02Sensitive Information Disclosure

MITRE ATLAS

AML.T0024.000Infer Training Data Membership

Suggested from the entry's categories.

Cite

@misc{tong2026pretraining,
  title = {{Pretraining Data Exposure in Large Language Models: A Survey of Membership Inference, Data Contamination, and Security Implications}},
  author = {Ziyi Tong and Feifei Sun and Le Minh Nguyen},
  year = {2026},
  month = may,
  eprint = {2605.26133},
  archivePrefix = {arXiv},
  doi = {10.1007/978-3-031-97144-0_14},
  url = {https://arxiv.org/abs/2605.26133}
}