4 papers · 1 filter
AutoPrompt: Automated Red-Teaming of Text-to-Image Models via LLM-Driven Adversarial Prompts
Yufan Liu, Wanqian Zhang, Huashan Chen +4
Despite rapid advancements in text-to-image (T2I) models, their safety mechanisms are vulnerable to adversarial prompts, which maliciously generate unsafe images. Current red-teami…
Prediction Exposes Your Face: Black-box Model Inversion via Prediction Alignment
Yufan Liu, Wanqian Zhang, Dayan Wu +3
Model inversion (MI) attack reconstructs the private training data of a target model given its output, posing a significant threat to deep learning models and data privacy. On one…
Towards Flexible Evaluation for Generative Visual Question Answering
Huishan Ji, Qingyi Si, Zheng Lin +1
Throughout rapid development of multimodal large language models, a crucial ingredient is a fair and accurate evaluation of their multimodal comprehension abilities. Although Visua…
Disrupting Diffusion: Token-Level Attention Erasure Attack against Diffusion-based Customization
Yisu Liu, Jinyang An, Wanqian Zhang +4
With the development of diffusion-based customization methods like DreamBooth, individuals now have access to train the models that can generate their personalized images. Despite…