TY - GEN
T1 - Fight Fire with Fire
T2 - 2021 Findings of the Association for Computational Linguistics, Findings of ACL: EMNLP 2021
AU - Wullach, Tomer
AU - Adler, Amir
AU - Minkov, Einat
N1 - Publisher Copyright: © 2021 Association for Computational Linguistics.
PY - 2021
Y1 - 2021
N2 - Automatic hate speech detection is hampered by the scarcity of labeled datasetd, leading to poor generalization. We employ pretrained language models (LMs) to alleviate this data bottleneck. We utilize the GPT LM for generating large amounts of synthetic hate speech sequences from available labeled examples, and leverage the generated data in fine-tuning large pretrained LMs on hate detection. An empirical study using the models of BERT, RoBERTa and ALBERT, shows that this approach improves generalization significantly and consistently within and across data distributions. In fact, we find that generating relevant labeled hate speech sequences is preferable to using out-of-domain, and sometimes also within-domain, human-labeled examples.
AB - Automatic hate speech detection is hampered by the scarcity of labeled datasetd, leading to poor generalization. We employ pretrained language models (LMs) to alleviate this data bottleneck. We utilize the GPT LM for generating large amounts of synthetic hate speech sequences from available labeled examples, and leverage the generated data in fine-tuning large pretrained LMs on hate detection. An empirical study using the models of BERT, RoBERTa and ALBERT, shows that this approach improves generalization significantly and consistently within and across data distributions. In fact, we find that generating relevant labeled hate speech sequences is preferable to using out-of-domain, and sometimes also within-domain, human-labeled examples.
UR - https://www.scopus.com/pages/publications/85129197068
U2 - 10.18653/v1/2021.findings-emnlp.402
DO - 10.18653/v1/2021.findings-emnlp.402
M3 - Conference contribution
T3 - Findings of the Association for Computational Linguistics, Findings of ACL: EMNLP 2021
SP - 4699
EP - 4705
BT - Findings of the Association for Computational Linguistics, Findings of ACL
A2 - Moens, Marie-Francine
A2 - Huang, Xuanjing
A2 - Specia, Lucia
A2 - Yih, Scott Wen-Tau
PB - Association for Computational Linguistics (ACL)
Y2 - 7 November 2021 through 11 November 2021
ER -