UoB at SemEval-2021 Task 5: Extending Pre-Trained Language Models to Include Task and Domain-Specific Information for Toxic Span Prediction

Erik Yan, Harish Tayyar Madabushi


Abstract
Toxicity is pervasive in social media and poses a major threat to the health of online communities. The recent introduction of pre-trained language models, which have achieved state-of-the-art results in many NLP tasks, has transformed the way in which we approach natural language processing. However, the inherent nature of pre-training means that they are unlikely to capture task-specific statistical information or learn domain-specific knowledge. Additionally, most implementations of these models typically do not employ conditional random fields, a method for simultaneous token classification. We show that these modifications can improve model performance on the Toxic Spans Detection task at SemEval-2021 to achieve a score within 4 percentage points of the top performing team.
Anthology ID:
2021.semeval-1.28
Volume:
Proceedings of the 15th International Workshop on Semantic Evaluation (SemEval-2021)
Month:
August
Year:
2021
Address:
Online
Venues:
ACL | IJCNLP | SemEval
SIG:
SIGLEX
Publisher:
Association for Computational Linguistics
Note:
Pages:
243–248
Language:
URL:
https://aclanthology.org/2021.semeval-1.28
DOI:
10.18653/v1/2021.semeval-1.28
Bibkey:
Copy Citation:
PDF:
https://aclanthology.org/2021.semeval-1.28.pdf
Code
 erikdyan/toxic_span_detection