相关论文: No Caption, No Problem: Caption-Free Membership In…

Large-scale pretrained vision-language models like CLIP have demonstrated remarkable zero-shot image classification capabilities across diverse domains. To enhance CLIP's performance while preserving the zero-shot paradigm, various…

The pervasive deployment of deep learning models across critical domains has concurrently intensified privacy concerns due to their inherent propensity for data memorization. While Membership Inference Attacks (MIAs) serve as the gold…

Fine-tuned language models pose significant privacy risks, as they may memorize and expose sensitive information from their training data. Membership inference attacks (MIAs) provide a principled framework for auditing these risks, yet…

Large-scale vision-language models (VLMs) such as CLIP exhibit strong zero-shot generalization, but adapting them to downstream tasks typically requires costly labeled data. Existing unsupervised self-training methods rely on…

Vision-language pretraining on large datasets of images-text pairs is one of the main building blocks of current Vision-Language Models. While with additional training, these models excel in various downstream tasks, including visual…

The creation of high-quality human-labeled image-caption datasets presents a significant bottleneck in the development of Visual-Language Models (VLMs). In this work, we investigate an approach that leverages the strengths of Large Language…

Safety classifiers are essential safeguards within generative AI systems, filtering harmful content or identifying at-risk users when interacting with large language models. Despite their necessity, these models are trained on sensitive…

Membership inference attacks (MIAs) aim to determine whether a sample was part of a model's training set, posing serious privacy risks for modern machine-learning systems. Existing MIAs primarily rely on static indicators, such as loss or…

Deep learning models, while achieving remarkable performances, are vulnerable to membership inference attacks (MIAs). Although various defenses have been proposed, there is still substantial room for improvement in the privacy-utility…

Determining whether a dataset was part of a machine learning model's training data pool can reveal privacy vulnerabilities, a challenge often addressed through membership inference attacks (MIAs). Traditional MIAs typically require access…

Masked Image Modeling (MIM) has achieved significant success in the realm of self-supervised learning (SSL) for visual recognition. The image encoder pre-trained through MIM, involving the masking and subsequent reconstruction of input…

Recent progress with conditional image diffusion models has been stunning, and this holds true whether we are speaking about models conditioned on a text description, a scene layout, or a sketch. Unconditional image diffusion models are…

Membership inference attacks (MIAs) aim to determine whether a specific data point was part of a model's training set, serving as effective tools for evaluating privacy leakage of vision models. However, existing MIAs implicitly assume…

Large unimodal foundation models for vision and language encode rich semantic structures, yet aligning them typically requires computationally intensive multimodal fine-tuning. Such approaches depend on large-scale parameter updates, are…

The vulnerability of machine learning models to Membership Inference Attacks (MIAs) has garnered considerable attention in recent years. These attacks determine whether a data sample belongs to the model's training set or not. Recent…

Open-vocabulary semantic segmentation aims at segmenting arbitrary categories expressed in textual form. Previous works have trained over large amounts of image-caption pairs to enforce pixel-level multimodal alignments. However, captions…

Pretrained Vision-Language-Action (VLA) policies have achieved strong single-step manipulation, but their inference remains largely memoryless, which is brittle in non-Markovian long-horizon settings with occlusion, state aliasing, and…

Recent work in the privacy literature shows that sample-targeted membership inference attacks (MIAs) significantly outperform untargeted approaches by a wide margin. Motivated by this observation, we address the following question: can the…

All prior membership inference attacks for fine-tuned language models use hand-crafted heuristics (e.g., loss thresholding, Min-K\%, reference calibration), each bounded by the designer's intuition. We introduce the first transferable…

We argue that generative text-to-image models often struggle with prompt adherence due to the noisy and unstructured nature of large-scale datasets like LAION-5B. This forces users to rely heavily on prompt engineering to elicit desirable…