Context-Aware Membership Inference Attacks against Pre-trained Large Language Models
This addresses security and privacy concerns for users of pre-trained LLMs, though it is an incremental advancement in attack techniques.
The paper tackles the problem of membership inference attacks on pre-trained large language models by adapting statistical tests to perplexity dynamics of subsequences, achieving significant performance improvements over prior methods.
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.