Hate speech propagates negativity and divisiveness online, undermining inclusivity. It is acknowledged that counterspeech can help lessen these negative effects. Despite its significance in changing attitudes, little study has looked at how haters respond to counterspeech, despite several studies examining user-generated counterspeech on social media. In order to determine if counterspeech causes the hater to return the debate and whether this reentry is hateful, this study examines counterspeech from the viewpoint of the hater. To evaluate the effect of counterspeech, we aggregate the Reddit Echoes of Hate dataset (ReEco) with triple-turn discussions containing the responses of haters. Linguistic study sheds light on how haters react differently to counterspeech terminology. The 3-way classification model works better than the two-stage response predictor, which predicts reentry first and then identifies the kind of reentry, according to experimental data. An evaluation of the most frequent mistakes found by the top-performing model marks the study’s conclusion.

https://arxiv.org/abs/2501.16235

By author

Leave a Reply

Your email address will not be published. Required fields are marked *