4 papers
Capacity Overflow: A Blind Spot for Backdoor Attacks in Vision MoE
Xiaocheng Zou, Tiancheng Zheng, Xiaolin Xu +1
Mixture-of-Experts (MoE) has become a prevalent paradigm for scaling Vision Transformers efficiently. To ensure computational scalability and prevent expert overload, Vision MoE ar…
ConQuER: Modular Architectures for Control and Bias Mitigation in IQP Quantum Generative Models
Xiaocheng Zou, Shijin Duan, Charles Fleming +4
Quantum generative models based on instantaneous quantum polynomial (IQP) circuits show great promise in learning complex distributions while maintaining classical trainability. Ho…
MetaSeal: Defending Against Image Attribution Forgery Through Content-Dependent Cryptographic Watermarks
Tong Zhou, Ruyi Ding, Gaowen Liu +5
The rapid growth of digital and AI-generated images has amplified the need for secure and verifiable methods of image attribution. While digital watermarking offers more robust pro…
ProDiF: Protecting Domain-Invariant Features to Secure Pre-Trained Models Against Extraction
Tong Zhou, Shijin Duan, Gaowen Liu +4
Pre-trained models are valuable intellectual property, capturing both domain-specific and domain-invariant features within their weight spaces. However, model extraction attacks th…