State-of-the-art approaches for hate-speech detection usually exhibit poor performance in out-of-domain settings. This occurs, typically, due to classifiers overemphasizing source-specific information that negatively impacts its domain invariance. Prior work has attempted to penalize terms related to hate-speech from manually curated lists using feature attribution methods, which quantify the importance assigned to input terms by the classifier when making a prediction. We, instead, propose a domain adaptation approach that automatically extracts and penalizes source-specific terms using a domain classifier, which learns to differentiate between domains, and feature-attribution scores for hate-speech classes, yielding consistent improvements in cross-domain evaluation.
@article{arxiv.2209.08681,
title = {Domain Classification-based Source-specific Term Penalization for Domain Adaptation in Hate-speech Detection},
author = {Tulika Bose and Nikolaos Aletras and Irina Illina and Dominique Fohr},
journal= {arXiv preprint arXiv:2209.08681},
year = {2022}
}