4 papers
PAR-AdvGAN: Improving Adversarial Attack Capability with Progressive Auto-Regression AdvGAN
Jiayu Zhang, Zhiyu Zhu, Xinyi Wang +4
Deep neural networks have demonstrated remarkable performance across various domains. However, they are vulnerable to adversarial examples, which can lead to erroneous predictions.…
ABE: A Unified Framework for Robust and Faithful Attribution-Based Explainability
Zhiyu Zhu, Jiayu Zhang, Zhibo Jin +2
Attribution algorithms are essential for enhancing the interpretability and trustworthiness of deep learning models by identifying key features driving model decisions. Existing fr…
Narrowing Information Bottleneck Theory for Multimodal Image-Text Representations Interpretability
Zhiyu Zhu, Zhibo Jin, Jiayu Zhang +4
The task of identifying multimodal image-text representations has garnered increasing attention, particularly with models such as CLIP (Contrastive Language-Image Pretraining), whi…
Attribution for Enhanced Explanation with Transferable Adversarial eXploration
Zhiyu Zhu, Jiayu Zhang, Zhibo Jin +3
The interpretability of deep neural networks is crucial for understanding model decisions in various applications, including computer vision. AttEXplore++, an advanced framework bu…