Paper: "Black-box Generation of Adversarial Text Sequences to Evade Deep Learning Classifiers", at 2018 IEEE Security and Privacy Workshops (SPW), co-located with the 39th IEEE Symposium on Security and Privacy) - View it on GitHub