mlsanityData to 7 Oct 2026

The Phantom Menace: Unmasking Privacy Leakages in Vision-Language Models

arXiv 2408.01228 ↗First posted 2 Aug 2024cs.CVcited by 8

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

Abstract (arXiv)

Vision-Language Models (VLMs) combine visual and textual understanding, rendering them well-suited for diverse tasks like generating image captions and answering visual questions across various domains. However, these capabilities are built upon training on large amount of uncurated data crawled from the web. The latter may include sensitive information that VLMs could memorize and leak, raising significant privacy concerns. In this paper, we assess whether these vulnerabilities exist, focusing on identity leakage. Our study leads to three key findings: (i) VLMs leak identity information, even when the vision-language alignment and the fine-tuning use anonymized data; (ii) context has little influence on identity leakage; (iii) simple, widely used anonymization techniques, like blurring, are not sufficient to address the problem. These findings underscore the urgent need for robust privacy protection strategies when deploying VLMs. Ethical awareness and responsible development practices are essential to mitigate these risks.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Placed in MLLM jailbreaks and adversarial robustness, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 32 of its 46 references that are in the map. First posted on arXiv 2 Aug 2024.

Builds on

32 of 46 references.

References · 32

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  3. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  4. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  5. 2023Segment Anythingcited by 8,158
  6. 2023Visual Instruction Tuningcited by 7,792
  7. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,592
  8. 2023BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Modelscited by 6,069
  9. 2023Improved Baselines with Visual Instruction Tuningcited by 4,362
  10. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,131
  11. 2021Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervisioncited by 3,848
  12. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,464
20 more
  1. 2023Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detectioncited by 3,274
  2. 2023Sigmoid Loss for Language Image Pre-Trainingcited by 3,048
  3. 2023MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Modelscited by 2,689
  4. 2015VQA: Visual Question Answeringcited by 2,489
  5. 2016Learning without Forgettingcited by 2,465
  6. 2024Mixtral of Expertscited by 1,785
  7. 2021Conceptual 12M: Pushing Web-Scale Image-Text Pre-Training To Recognize Long-Tail Visual Conceptscited by 1,209
  8. 2017Memory Aware Synapses: Learning what (not) to forgetcited by 1,094
  9. 2024PaliGemma: A versatile 3B VLM for transfercited by 679
  10. 2018A Corpus for Reasoning About Natural Language Grounded in Photographscited by 457
  11. 2018Scalable Private Learning with PATEcited by 266
  12. 2021Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Conceptscited by 255
  13. 2024Leak, Cheat, Repeat: Data Contamination and Evaluation Malpractices in Closed-Source LLMscited by 224
  14. 2018A Comprehensive Survey of Deep Learning for Image Captioningcited by 213
  15. 2023Can Neural Network Memorization Be Localized?cited by 66
  16. 2023Continual Learning: Applications and the Road Forwardcited by 59
  17. 2022Image-text Retrieval: A Survey on Recent Research and Developmentcited by 55
  18. 2023Unveiling Security, Privacy, and Ethical Concerns of ChatGPTcited by 53
  19. 2019AnonymousNet: Natural Face De-Identification with Measurable Privacycited by 50
  20. 2022Does CLIP Know My Face?cited by 19

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 1; 2025: 4; 2026: 3.

8 papers.

  1. 2024AutoTrust: Benchmarking Trustworthiness in Large Vision Language Models for Autonomous Drivingcited by 35
  2. 2025MobileA3gent: Training Mobile GUI Agents Using Decentralized Self-Sourced Data from Diverse Userscited by 10
  3. 2025FedMABench: Benchmarking Mobile Agents on Decentralized Heterogeneous User Datacited by 3
  4. 2025FedVLM: Scalable Personalized Vision-Language Models through Federated Learningcited by 2
  5. 2025OpenLVLM-MIA: A Controlled Benchmark Revealing the Limits of Membership Inference Attacks on Large Vision-Language Modelscited by 1
  6. 2026Differentially Private Multimodal In-Context Learningcited by 1
  7. 2026From Measurement to Mitigation: Quantifying and Reducing Identity Leakage in Image Representation Encoders with Linear Subspace Removalcited by 0
  8. 2026Seeing Without Exposing: Adaptive Privacy Control for Open-World, Context-Hungry MLLMscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Assessing Visual Privacy Risks in Multimodal AI: A Novel Taxonomy-Grounded Evaluation of Vision-Language Models
  2. 2024Privacy-Aware Visual Language Models
  3. 2025On the Robustness of Vision-Language Models in Zero-shot Privacy Classification
  4. 2024Private Attribute Inference from Images with Vision-Language Models
  5. 2025Defeating Cerberus: Concept-Guided Privacy-Leakage Mitigation in Multimodal Language Models
  6. 2024Multi-PA: A Multi-perspective Benchmark on Privacy Assessment for Large Vision-Language Models
  7. 2025VIP: Visual Information Protection through Adversarial Attacks on Vision-Language Models
  8. 2026PII-VisBench: Evaluating Personally Identifiable Information Safety in Vision Language Models Along a Continuum of Visibility