LLMs show improved directionality detection after fine-tuning but perform near chance at cyberbullying and anti-bullying classification, pointing to weak semantic understanding of informal social media language.
In: Proceedings of the 13th international workshop on semantic evaluation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Evaluating LLMs Capabilities Towards Understanding Social Dynamics
LLMs show improved directionality detection after fine-tuning but perform near chance at cyberbullying and anti-bullying classification, pointing to weak semantic understanding of informal social media language.