VL-BERT+: Detecting Protected Groups in Hateful Multimodal Memes

Piush Aggarwal, Michelle Espranita Liman, Darina Gold, Torsten Zesch


Abstract
This paper describes our submission (winning solution for Task A) to the Shared Task on Hateful Meme Detection at WOAH 2021. We build our system on top of a state-of-the-art system for binary hateful meme classification that already uses image tags such as race, gender, and web entities. We add further metadata such as emotions and experiment with data augmentation techniques, as hateful instances are underrepresented in the data set.
Anthology ID:
2021.woah-1.22
Volume:
Proceedings of the 5th Workshop on Online Abuse and Harms (WOAH 2021)
Month:
August
Year:
2021
Address:
Online
Venues:
ACL | IJCNLP | WOAH
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
207–214
Language:
URL:
https://aclanthology.org/2021.woah-1.22
DOI:
10.18653/v1/2021.woah-1.22
Bibkey:
Cite (ACL):
Piush Aggarwal, Michelle Espranita Liman, Darina Gold, and Torsten Zesch. 2021. VL-BERT+: Detecting Protected Groups in Hateful Multimodal Memes. In Proceedings of the 5th Workshop on Online Abuse and Harms (WOAH 2021), pages 207–214, Online. Association for Computational Linguistics.
Cite (Informal):
VL-BERT+: Detecting Protected Groups in Hateful Multimodal Memes (Aggarwal et al., WOAH 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.woah-1.22.pdf
Data
Hateful Memes