Automatic Calibration for Membership Inference Attack on Large Language Models
Preprint
Abstract
Membership Inference Attacks (MIAs) have recently been employed to determine whether a specific text was part of the pre-training data of Large Language Models (LLMs). However, existing methods often misinfer non-members as members, leading to a high false positive rate, or depend on additional reference models for probability calibration, which limits their practicality. To overcome these challenges, we introduce a novel framework called Automatic Calibration Membership Inference Attack (ACMIA), which utilizes a tunable temperature to calibrate output probabilities effectively. This approach is inspired by our theoretical insights into maximum likelihood estimation during the pre-training of LLMs. We introduce ACMIA in three configurations designed to accommodate different levels of model access and increase the probability gap between members and non-members, improving the reliability and robustness of membership inference. Extensive experiments on various open-source LLMs demonstrate that our proposed attack is highly effective, robust, and generalizable, surpassing state-of-the-art baselines across three widely used benchmarks. Our code is available at: \href{https://github.com/Salehzz/ACMIA}{\textcolor{blue}{Github}}.Keywords
All Related Versions
This publication has 11 references indexed in Scilit:
- Generative Large Language Model—Powered Conversational AI App for Personalized Risk Assessment: Case Study in COVID-19JMIR AI, 2025
- Rethinking machine unlearning for large language modelsNature Machine Intelligence, 2025
- Pretraining Data Detection for Large Language Models: A Divergence-based Calibration Method2024
- Probing Language Models for Pre-training Data Detection2024
- Unveiling security, privacy, and ethical concerns of ChatGPTJournal of Information and Intelligence, 2023
- Privacy Auditing with One (1) Training Run2023
- Emergent and Predictable Memorization in Large Language Models2023
- Membership Inference Attacks against Language Models via Neighbourhood Comparison2023
- Quantifying Privacy Risks of Masked Language Models Using Membership Inference Attacks2022
- GPT-NeoX-20B: An Open-Source Autoregressive Language Model2022