Text Toxicity Level Detection using Deep Contextualized Embedding Models
Summary
Deep contextualized embedding models applied to grading the toxicity level of text, framing moderation as a graded rather than binary classification problem.
Topics
NLPContent ModerationTransformers
More research
See all Other papers.
Working on something adjacent?
I'm open to research collaboration on LLM adaptation, low-resource NLP, and clinical machine learning.