---
title: "PAR-AdvGAN: Improving Adversarial Attack Capability with Progressive Auto-Regression AdvGAN (Extended Abstract)"
authors:
  - "Jiayu Zhang"
  - "Zhiyu Zhu"
  - "Xinyi Wang"
  - "Silin Liao"
  - "Zhibo Jin"
  - "Flora Salim"
  - "Huaming Chen"
doi: "10.24963/ijcai.2026/930"
url: "https://ijcai.org/proceedings/2026/930"
pdf: "https://ijcai.org/proceedings/2026/0930.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "8301-8305"
track: "Sister Conferences Best Papers"
keywords:
  - "Machine Learning: Trustworthy machine learning"
  - "AI Ethics, Trust, Fairnes: Trustworthy AI"
  - "AI Ethics, Trust, Fairnes: Safety and robustness"
---

# PAR-AdvGAN: Improving Adversarial Attack Capability with Progressive Auto-Regression AdvGAN (Extended Abstract)

Jiayu Zhang, Zhiyu Zhu, Xinyi Wang, Silin Liao, Zhibo Jin, Flora Salim, Huaming Chen

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 8301-8305. DOI: [10.24963/ijcai.2026/930](https://doi.org/10.24963/ijcai.2026/930)

## Abstract

Deep neural networks are vulnerable to adversarial examples, posing serious security risks. While GAN-based attacks such as AdvGAN enable fast adversarial example generation, they typically operate in a single step, limiting attack effectiveness and transferability. In this work, we propose PAR-AdvGAN, a progressive auto-regressive GAN framework that iteratively refines adversarial perturbations. By conditioning each iteration on the previous output, PAR-AdvGAN produces stronger, more transferable attacks while maintaining low visual distortion and high generation speed. Experiments on multiple models and datasets demonstrate that PAR-AdvGAN significantly outperforms baseline methods in attack success rate, achieving up to 335.5 frames per second, highlighting its practical efficiency for black-box attack scenarios.