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.
|Title of host publication||Proceedings of the 15th International Workshop on Semantic Evaluation (SemEval-2021)|
|Place of Publication||Online|
|Publisher||Association for Computational Linguistics|
|Number of pages||6|
|Publication status||Published - 1 Aug 2021|