mlsanityData to 14 Oct 2026

When Visual Privacy Protection Meets Multimodal Large Language Models

arXiv 2603.13978 ↗First posted 14 Mar 2026cs.CVcited by 0

On the map: Multimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness

Abstract (arXiv)

The emergence of Multimodal Large Language Models (MLLMs) and the widespread usage of MLLM cloud services such as GPT-4V raised great concerns about privacy leakage in visual data. As these models are typically deployed in cloud services, users are required to submit their images and videos, posing serious privacy risks. However, how to tackle such privacy concerns is an under-explored problem. Thus, in this paper, we aim to conduct a new investigation to protect visual privacy when enjoying the convenience brought by MLLM services. We address the practical case where the MLLM is a "black box", i.e., we only have access to its input and output without knowing its internal model information. To tackle such a challenging yet demanding problem, we propose a novel framework, in which we carefully design the learning objective with Pareto optimality to seek a better trade-off between visual privacy and MLLM's performance, and propose critical-history enhanced optimization to effectively optimize the framework with the black-box MLLM. Our experiments show that our method is effective on different benchmarks.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in MLLM jailbreaks and adversarial robustness, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 21 of its 49 references that are in the map. First posted on arXiv 14 Mar 2026.

Builds on

21 of 49 references.

References · 21

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  3. 2023GPT-4 Technical Reportcited by 18,396
  4. 2023Visual Instruction Tuningcited by 7,869
  5. 2023MMBench: Is Your Multi-modal Model an All-around Player?cited by 2,141
  6. 2023Video-LLaVA: Learning United Visual Representation by Alignment Before Projectioncited by 1,402
  7. 2019OK-VQA: A Visual Question Answering Benchmark Requiring External Knowledgecited by 1,121
  8. 2023Video-ChatGPT: Towards Detailed Video Understanding via Large Vision and Language Modelscited by 1,062
  9. 2023A Survey on Multimodal Large Language Modelscited by 909
  10. 2023MiniGPT-v2: large language model as a unified interface for vision-language multi-task learningcited by 625
  11. 2016Delving into Transferable Adversarial Examples and Black-box Attackscited by 536
  12. 2021Deep Learning on a Data Diet: Finding Important Examples Early in Trainingcited by 533
9 more
  1. 2023Scalable Extraction of Training Data from (Production) Language Modelscited by 434
  2. 2017Towards a Visual Privacy Advisor: Understanding and Predicting Privacy Risks in Imagescited by 101
  3. 2023BlackVIP: Black-Box Visual Prompting for Robust Transfer Learningcited by 70
  4. 2023Unveiling Security, Privacy, and Ethical Concerns of ChatGPTcited by 53
  5. 2022SPAct: Self-supervised Privacy Preservation for Action Recognitioncited by 47
  6. 2024A Survey on Evaluation of Multimodal Large Language Modelscited by 46
  7. 2023TeD-SPAD: Temporal Distinctiveness for Self-supervised Privacy-preservation for video Anomaly Detectioncited by 23
  8. 2024Selective, Interpretable, and Motion Consistent Privacy Attribute Obfuscation for Action Recognitioncited by 8
  9. 2021Memory Augmented Optimizers for Deep Learningcited by 5

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024Multi-PA: A Multi-perspective Benchmark on Privacy Assessment for Large Vision-Language Models
  2. 2026Seeing Without Exposing: Adaptive Privacy Control for Open-World, Context-Hungry MLLMs
  3. 2025Assessing Visual Privacy Risks in Multimodal AI: A Novel Taxonomy-Grounded Evaluation of Vision-Language Models
  4. 2025Defeating Cerberus: Concept-Guided Privacy-Leakage Mitigation in Multimodal Language Models
  5. 2026Unveiling Privacy Risks in Multi-modal Large Language Models: Task-specific Vulnerabilities and Mitigation Challenges
  6. 2025On the Robustness of Vision-Language Models in Zero-shot Privacy Classification
  7. 2025When Privacy Meets Recovery: The Overlooked Half of Surrogate-Driven Privacy Preservation for MLLM Editing
  8. 2024The Phantom Menace: Unmasking Privacy Leakages in Vision-Language Models