Closed DavidHerel closed 1 year ago
Hi David, thanks for your contribution! I take a quick read and think it is a great and timely contribution as you conduct adversarial example re-evaluation on semantic preserving under real-world application (hate speech detection) and find that most adversarial examples generated should be discarded. I propose listing your paper under the Benchmark and Evaluation
section for better exposure to people working in the textual adversarial domain as other papers
are mainly for relevant papers in textual adversarial domain
(e.g., multi-modal adversarial examples).
Thank you for the response! I have moved it to the suggested section.
Thanks! I will merge your PR now. Great works!
Add our ECAI 2023 paper: "Preserving Semantics in Textual Adversarial Attacks".