Saved in:
Bibliographic Details
Main Authors: Tong, Ziyi, Sun, Feifei, Nguyen, Le Minh
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.03121
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913151563333632
author Tong, Ziyi
Sun, Feifei
Nguyen, Le Minh
author_facet Tong, Ziyi
Sun, Feifei
Nguyen, Le Minh
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2512_03121
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Lost in Modality: Evaluating the Effectiveness of Text-Based Membership Inference Attacks on Large Multimodal Models
Tong, Ziyi
Sun, Feifei
Nguyen, Le Minh
Cryptography and Security
Artificial Intelligence
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.
title Lost in Modality: Evaluating the Effectiveness of Text-Based Membership Inference Attacks on Large Multimodal Models
topic Cryptography and Security
Artificial Intelligence
url https://arxiv.org/abs/2512.03121