Offense and Denfense

Offense

Computer Vision

[2018 AAAI] EAD: Elastic-Net Attacks to Deep Neural Networks via Adversarial Examples, [paper], [bibtex], sources: [IBM/EAD-Attack], [ysharma1126/EAD_Attack].

Natural Language Processing

[2018 SPW] Black-box Generation of Adversarial Text Sequences to Evade Deep Learning Classifiers, [paper], [bibtex], sources: [QData/deepWordBug].
[2019 NDSS] TextBugger: Generating Adversarial Text Against Real-world Applications, [paper], [bibtex].

Defense

Computer Vision

[2018 ICLR] Defense-GAN: Protecting Classifier Against Adversarial Attacks Using Generative Models, [paper], [bibtex], sources: [kabkabm/defensegan].
[2019 ICLR] Improving the Generalization of Adversarial Training with Domain Adaptation, [paper], [bibtex], sources: [JHL-HUST/ATDA].

Natural Language Processing