Original paper
Trick Me If You Can: Human-in-the-Loop Generation of Adversarial Examples for Question Answering
Volume: 7, Pages: 387 - 401
Published: Jul 26, 2019
Abstract
Adversarial evaluation stress-tests a model’s understanding of natural language. Because past approaches expose superficial patterns, the resulting adversarial examples are limited in complexity and diversity. We propose human- in-the-loop adversarial generation, where human authors are guided to break models. We aid the authors with interpretations of model predictions through an interactive user interface. We apply this generation framework to...
Paper Details
Title
Trick Me If You Can: Human-in-the-Loop Generation of Adversarial Examples for Question Answering
Published Date
Jul 26, 2019
Volume
7
Pages
387 - 401