Detecting Training Data For Large Language Models: A Survey

Chen Yang

doi:10.1145/3779430

2026-05-23

PRE

DOI：10.1145/3779430

原文求助

摘要

En 中文

随着大型语言模型（LLMs）的不断发展，其训练所使用的数据范围和多样性正显著扩大。然而，LLMs的训练数据集可能不可避免地包含敏感信息，如个人数据或版权材料，若模型生成与这些来源高度相似或完全相同的文本，将导致隐私泄露或版权侵权风险。这促使人们关注是否检测文本数据被用于LLMs训练的问题。迄今为止，关于检测人工智能（AI）模型训练数据使用的研究主要集中于传统机器学习（ML）模型，而针对LLMs的研究仍相对不成熟。对该领域研究进展的缺乏了解阻碍了更有效检测方法的发展。因此，本文旨在填补这一空白，通过分析LLMs训练数据的检测方法。具体而言，我们分析了可提供给检测器的LLMs信息、主要检测方法、判定指标，并讨论了该领域的技术挑战及未来研究的潜在方向。

Keyword:

Large language models

detecting training data

期刊

ACM Computing Surveys

IF:

论文数:

2.5K

被引数:

3.5W

机构

university of chinese academy of sciences, cas

学者数:

4.1W

论文数: 3.8W

被引数: 75

university of hong kong

学者数:

3.6K

论文数: 1.7K

被引数: 0

beijing institute of technology

学者数:

5.5W

论文数: 4.0W

被引数: 63

chinese academy of sciences

学者数:

56.7W

论文数: 45.0W

被引数: 704

查看更多机构

引用论文

Machine Learning with Membership Privacy using Adversarial Regularization

Proceedings of the 2018 ACM SIGSAC Conference on Computer and Communications Security

IF0

2018-10-15

OAAI

Milad Nasr; Reza Shokri; Amir Houmansadr

Paraphrase Generation: A Survey of the State of the Art

Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing

IF0

2021-01-01

OAAI

Jianing Zhou; Suma Bhat

Machine Unlearning: A Survey 机器学习: 一项调查

ACM COMPUTING SURVEYS

IF28

2023-08-28

OAAI

Xu, Heng; Zhu, Tianqing; Zhang, Lefeng; Zhou, Wanlei; Yu, Philip S.

Knowledge Distillation: A Survey 知识蒸馏: 一项调查

INTERNATIONAL JOURNAL OF COMPUTER VISION

IF9.3

2021-03-22

1.5K

PREAI

Gou, Jianping; Yu, Baosheng; Maybank, Stephen J.; Tao, Dacheng

A Survey of Text Similarity Approaches

International Journal of Computer Applications

IF0

2013-04-18

OAAI

Wael H.Gomaa; Aly A. Fahmy

Right to Be Forgotten in the Age of Machine Learning

Advances in Intelligent Systems and Computing

IF0

2021-03-15

PREAI

Quang-Vinh Dang

Information-Guided Identification of Training Data Imprint in (Proprietary) Large Language Models

Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)

IF0

2025-01-01

PREAI

Ravichander,Abhilasha; Fisher,Jillian; Sorensen,Taylor; Lu,Ximing; Antoniak,Maria; Lin,Bill Yuchen; Mireshghallah,Niloofar; Bhagavatula,Chandra; Choi,Yejin

Membership Inference Attacks Against Machine Learning Models 针对机器学习模型的成员推理攻击

2017 IEEE Symposium on Security and Privacy (SP)

IF0

2017-05-01

OAAI

Reza Shokri; Marco Stronati; Congzheng Song; Vitaly Shmatikov

The performance of text similarity algorithms

International Journal of Advances in Intelligent Informatics

IF0

2018-03-31

OAAI

Didik Dwi Prasetya; Aji Prasetya Wibawa; Tsukasa Hirashima