A recipe for watermarking diffusion models

Y Zhao, T Pang, C Du, X Yang, NM Cheung… - arxiv preprint arxiv …, 2023‏ - arxiv.org
Diffusion models (DMs) have demonstrated advantageous potential on generative tasks.
Widespread interest exists in incorporating DMs into downstream applications, such as …

Task-specific fine-tuning via variational information bottleneck for weakly-supervised pathology whole slide image classification

H Li, C Zhu, Y Zhang, Y Sun, Z Shui… - Proceedings of the …, 2023‏ - openaccess.thecvf.com
Abstract While Multiple Instance Learning (MIL) has shown promising results in digital
Pathology Whole Slide Image (WSI) analysis, such a paradigm still faces performance and …

Re-thinking model inversion attacks against deep neural networks

NB Nguyen, K Chandrasegaran… - Proceedings of the …, 2023‏ - openaccess.thecvf.com
Abstract Model inversion (MI) attacks aim to infer and reconstruct private training data by
abusing access to a model. MI attacks have raised concerns about the leaking of sensitive …

Image synthesis under limited data: A survey and taxonomy

M Yang, Z Wang - International Journal of Computer Vision, 2025‏ - Springer
Deep generative models, which target reproducing the data distribution to produce novel
images, have made unprecedented advancements in recent years. However, one critical …

Exploring incompatible knowledge transfer in few-shot image generation

Y Zhao, C Du, M Abdollahzadeh… - Proceedings of the …, 2023‏ - openaccess.thecvf.com
Few-shot image generation (FSIG) learns to generate diverse and high-fidelity images from
a target domain using a few (eg, 10) reference samples. Existing FSIG methods select …

Label-only model inversion attacks via knowledge transfer

BN Nguyen, K Chandrasegaran… - Advances in …, 2023‏ - proceedings.neurips.cc
In a model inversion (MI) attack, an adversary abuses access to a machine learning (ML)
model to infer and reconstruct private training data. Remarkable progress has been made in …

Dss-net: Dynamic self-supervised network for video anomaly detection

P Wu, W Wang, F Chang, C Liu… - IEEE Transactions on …, 2023‏ - ieeexplore.ieee.org
Video Anomaly detection, aiming to detect the abnormal behaviors in surveillance videos, is
a challenging task since the anomalous events are diversified and complicated in different …

A survey on generative modeling with limited data, few shots, and zero shot

M Abdollahzadeh, T Malekzadeh, CTH Teo… - arxiv preprint arxiv …, 2023‏ - arxiv.org
In machine learning, generative modeling aims to learn to generate new data statistically
similar to the training data distribution. In this paper, we survey learning generative models …

Lfs-gan: Lifelong few-shot image generation

J Seo, JS Kang, GM Park - Proceedings of the IEEE/CVF …, 2023‏ - openaccess.thecvf.com
We address a challenging lifelong few-shot image generation task for the first time. In this
situation, a generative model learns a sequence of tasks using only a few samples per task …

Domain re-modulation for few-shot generative domain adaptation

Y Wu, Z Li, C Wang, H Zheng… - Advances in Neural …, 2023‏ - proceedings.neurips.cc
In this study, we delve into the task of few-shot Generative Domain Adaptation (GDA), which
involves transferring a pre-trained generator from one domain to a new domain using only a …