Adversarial Examples for Evaluating Reading Comprehension Systems

20171,277 citationsJournal Articlegold Open Access

Authors

Percy Liang · Laboratoire d'Informatique de Paris-Nord

Abstract

Standard accuracy metrics indicate that reading comprehension systems are making rapid progress, but the extent to which these systems truly understand language remains unclear.

Topics & Keywords

Topic Modeling Adversarial Robustness in Machine Learning Hate Speech and Cyberbullying Detection

UN Sustainable Development Goals

Quality Education

Publication Details

DOI: 10.18653/v1/d17-1215

Field-Weighted Citation Impact: 112.80