LeakyCLIP: Extracting Training Data from CLIP
By: Yunhao Chen , Shujie Wang , Xin Wang and more
Potential Business Impact:
Steals private pictures from AI's memory.
Understanding the memorization and privacy leakage risks in Contrastive Language--Image Pretraining (CLIP) is critical for ensuring the security of multimodal models. Recent studies have demonstrated the feasibility of extracting sensitive training examples from diffusion models, with conditional diffusion models exhibiting a stronger tendency to memorize and leak information. In this work, we investigate data memorization and extraction risks in CLIP through the lens of CLIP inversion, a process that aims to reconstruct training images from text prompts. To this end, we introduce \textbf{LeakyCLIP}, a novel attack framework designed to achieve high-quality, semantically accurate image reconstruction from CLIP embeddings. We identify three key challenges in CLIP inversion: 1) non-robust features, 2) limited visual semantics in text embeddings, and 3) low reconstruction fidelity. To address these challenges, LeakyCLIP employs 1) adversarial fine-tuning to enhance optimization smoothness, 2) linear transformation-based embedding alignment, and 3) Stable Diffusion-based refinement to improve fidelity. Empirical results demonstrate the superiority of LeakyCLIP, achieving over 358% improvement in Structural Similarity Index Measure (SSIM) for ViT-B-16 compared to baseline methods on LAION-2B subset. Furthermore, we uncover a pervasive leakage risk, showing that training data membership can even be successfully inferred from the metrics of low-fidelity reconstructions. Our work introduces a practical method for CLIP inversion while offering novel insights into the nature and scope of privacy risks in multimodal models.
Similar Papers
LeakyCLIP: Extracting Training Data from CLIP
Cryptography and Security
Steals private pictures from AI's memory.
Contrastive vision-language learning with paraphrasing and negation
CV and Pattern Recognition
Teaches computers to understand words that change meaning.
uCLIP: Parameter-Efficient Multilingual Extension of Vision-Language Models with Unpaired Data
CV and Pattern Recognition
Helps computers understand pictures in many languages.