Online Library

VOX-Pol Blog Post.

VOX-Pol Blog Post.

With the spread of social networks and their unfortunate use for hate speech, automatic detection of the latter has become a pressing problem. In this paper, we reproduce seven state-of-the-art hate speech detection models from prior work, and show that they perform well only when tested on the same type of data they were trained on. Based on these results, we argue that for successful hate speech detection, model architecture is less important than the type of data and labeling criteria. We further show that all proposed detection techniques are brittle against adversaries who can (automatically) insert typos, change word boundaries or add innocuous words to the original hate speech. A combination of these methods is also effective against Google Perspective – a cuttingedge solution from industry. Our experiments demonstrate that adversarial training does not completely mitigate the attacks, and using character-level features makes the models systematically more
attack-resistant than using word-level features.

With the spread of social networks and their unfortunate use for hate speech, automatic detection of the latter has become a pressing problem. In this paper, we reproduce seven state-of-the-art hate speech detection models from prior work, and show that they perform well only when tested on the same type of data they were trained on. Based on these results, we argue that for successful hate speech detection, model architecture is less important than the type of data and labeling criteria. We further show that all proposed detection techniques are brittle against adversaries who can (automatically) insert typos, change word boundaries or add innocuous words to the original hate speech. A combination of these methods is also effective against Google Perspective – a cutting edge solution from industry. Our experiments demonstrate that adversarial training does not completely mitigate the attacks, and using character-level features makes the models systematically more attack-resistant than using word-level features.

The rise of the alt-right as a potent and sometimes violent political force has been well documented. Yet the journey of an individual towards upholding these ideologies is less well understood. Alt-righters are not instantly converted, but rather incrementally nudged along a particular medial pathway. Drawing on video testimonies, chat logs, and other studies, this paper explores the interaction between this alt-right “pipeline” and the psyche of a user. It suggests three overlapping cognitive phases that occur within this journey: normalization, acclimation, and dehumanization. Finally, the article examines the individual who has reached the end of this journey, an extremist who nevertheless remains largely unregistered within traditional terrorist classifications.