Mike Laszkiewicz
YOU?
Author Swipe
View article: Lightweight Model Attribution and Detection of Synthetic Speech via Audio Residual Fingerprints
Lightweight Model Attribution and Detection of Synthetic Speech via Audio Residual Fingerprints Open
As speech generation technologies advance, so do risks of impersonation, misinformation, and spoofing. We present a lightweight, training-free approach for detecting synthetic speech and attributing it to its source model. Our method addre…
View article: Benchmarking the Fairness of Image Upsampling Methods
Benchmarking the Fairness of Image Upsampling Methods Open
Recent years have witnessed a rapid development of deep generative models for creating synthetic media, such as images and videos. While the practical applications of these models in everyday tasks are enticing, it is crucial to assess the…
View article: AnomalyDINO: Boosting Patch-based Few-shot Anomaly Detection with DINOv2
AnomalyDINO: Boosting Patch-based Few-shot Anomaly Detection with DINOv2 Open
Recent advances in multimodal foundation models have set new standards in few-shot anomaly detection. This paper explores whether high-quality visual features alone are sufficient to rival existing state-of-the-art vision-language models. …
View article: Benchmarking the Fairness of Image Upsampling Methods
Benchmarking the Fairness of Image Upsampling Methods Open
Recent years have witnessed a rapid development of deep generative models for creating synthetic media, such as images and videos. While the practical applications of these models in everyday tasks are enticing, it is crucial to assess the…
View article: Set-Membership Inference Attacks using Data Watermarking
Set-Membership Inference Attacks using Data Watermarking Open
In this work, we propose a set-membership inference attack for generative models using deep image watermarking techniques. In particular, we demonstrate how conditional sampling from a generative model can reveal the watermark that was inj…
View article: Single-Model Attribution of Generative Models Through Final-Layer Inversion
Single-Model Attribution of Generative Models Through Final-Layer Inversion Open
Recent breakthroughs in generative modeling have sparked interest in practical single-model attribution. Such methods predict whether a sample was generated by a specific generator or not, for instance, to prove intellectual property theft…
View article: Marginal Tail-Adaptive Normalizing Flows
Marginal Tail-Adaptive Normalizing Flows Open
Learning the tail behavior of a distribution is a notoriously difficult problem. By definition, the number of samples from the tail is small, and deep generative models, such as normalizing flows, tend to concentrate on learning the body o…
View article: Copula-Based Normalizing Flows
Copula-Based Normalizing Flows Open
Normalizing flows, which learn a distribution by transforming the data to samples from a Gaussian base distribution, have proven powerful density approximations. But their expressive power is limited by this choice of the base distribution…
View article: Thresholded Adaptive Validation: Tuning the Graphical Lasso for Graph Recovery
Thresholded Adaptive Validation: Tuning the Graphical Lasso for Graph Recovery Open
Many Machine Learning algorithms are formulated as regularized optimization problems, but their performance hinges on a regularization parameter that needs to be calibrated to each application at hand. In this paper, we propose a general c…