PrivMark: Pixel-Space Steering against Privacy Leakage in Vision-Language Models
cs.CV
Submitted: 2026-11-07
Updated: 2026-11-07
Terminology
Sources
- OpenFlamingo: An Open-Source Framework for Training Large Autoregressive Vision-Language Models
- Qwen2.5-VL Technical Report
- VISOR++: Universal Visual Inputs based Steering for Large Vision Language Models
- VIP: Visual Information Protection through Adversarial Attacks on Vision-Language Models
- Safe-LLaVA: A Privacy-Preserving Vision-Language Dataset and Benchmark for Biometric Safety
- On the Adversarial Robustness of Multi-Modal Foundation Models
- PII-VisBench: Evaluating Personally Identifiable Information Safety in Vision Language Models Along a Continuum of Visibility
- Leave My Images Alone: Preventing Multi-Modal Large Language Models from Analyzing Images via Visual Prompt Injection
- Steering Language Models With Activation Engineering
- InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models
- Representation Engineering: A Top-Down Approach to AI Transparency
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs