In a zero-shot benchmark on 5,080 YouTube comments, GPT-4.1 achieved the best F1 balance at 0.863, while Gemini favored recall and Claude favored precision.
Social media use and cyber-bullying: A cross-national analysis of young people in 42 countries,
1 Pith paper cite this work, alongside 359 external citations. Polarity classification is still indexing.
1
Pith paper citing it
359
external citations · OpenAlex
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Moderating Harm: Benchmarking Large Language Models for Cyberbullying Detection in YouTube Comments
In a zero-shot benchmark on 5,080 YouTube comments, GPT-4.1 achieved the best F1 balance at 0.863, while Gemini favored recall and Claude favored precision.