LLMs struggle to classify hateful events accurately but show promise with guidance

Original: Evaluating Large Language Models for Antisemitic Incident Classification

Source: arxiv.org ↗