Fine-grained taxonomic prompts substantially raise antisemitism-detection recall across four LLMs while lowering precision; larger conceptual resources add no further quantitative gain and post-Holocaust forms remain hardest.
An In-depth Analysis of Implicit and Subtle Hate Speech Messages
1 Pith paper cite this work, alongside 46 external citations. Polarity classification is still indexing.
1
Pith paper citing it
46
external citations · OpenAlex
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
You Frame It: How Conceptual Representations Shape LLM Detection and Reasoning about Antisemitism
Fine-grained taxonomic prompts substantially raise antisemitism-detection recall across four LLMs while lowering precision; larger conceptual resources add no further quantitative gain and post-Holocaust forms remain hardest.