Published in Under Review, 2024
This paper presents a study on extracting training data from personalized generative models using guidance.
Recommended citation: Wu X, Zhang J, Wu S. (2024). "Revealing the Unseen: Guiding Personalized Diffusion Models to Expose Training Data." arXiv preprint arXiv:2410.03039.
Download Paper
Published in Under Review, 2024
This paper explores diffusion models’ corruption stage in few-shot fine-tuning and proposes using Bayesian Neural Networks to mitigate image quality degradation.
Recommended citation: Wu X*, Zhang J*, Hua Y, et al. (2024). "Exploring Diffusion Models' Corruption Stage in Few-Shot Fine-Tuning and Mitigating with Bayesian Neural Networks." arXiv preprint arXiv:2405.19931. (Co-First Author)
Download Paper
Published in CVPR 2024, 2024
This paper presents a copyright authentication mechanism for diffusion models by analyzing the conceptual differences between fine-tuned and pretrained models.
Recommended citation: Wu X, Hua Y, Liang C, et al. (2024). "CGI-DM: Digital Copyright Authentication for Diffusion Models via Contrasting Gradient Inversion." CVPR 2024.
Download Paper
Published in ICML 2023 (Oral), 2023
This paper presents a framework to prevent painting imitation from diffusion models using adversarial examples.
Recommended citation: Liang C*, Wu X*, Hua Y, et al. (2023). "Adversarial Example Does Good: Preventing Painting Imitation from Diffusion Models via Adversarial Examples." ICML 2023 (Oral). (Co-First Author)
Download Paper