Score: 1

Lost in Modality: Evaluating the Effectiveness of Text-Based Membership Inference Attacks on Large Multimodal Models

Published: December 2, 2025 | arXiv ID: 2512.03121v1

By: Ziyi Tong, Feifei Sun, Le Minh Nguyen

Potential Business Impact:

Finds if private images were used to train AI.

Business Areas:
Text Analytics Data and Analytics, Software

Large Multimodal Language Models (MLLMs) are emerging as one of the foundational tools in an expanding range of applications. Consequently, understanding training-data leakage in these systems is increasingly critical. Log-probability-based membership inference attacks (MIAs) have become a widely adopted approach for assessing data exposure in large language models (LLMs), yet their effect in MLLMs remains unclear. We present the first comprehensive evaluation of extending these text-based MIA methods to multimodal settings. Our experiments under vision-and-text (V+T) and text-only (T-only) conditions across the DeepSeek-VL and InternVL model families show that in in-distribution settings, logit-based MIAs perform comparably across configurations, with a slight V+T advantage. Conversely, in out-of-distribution settings, visual inputs act as regularizers, effectively masking membership signals.

Page Count
6 pages

Category
Computer Science:
Cryptography and Security