Hybrid vocabulary-plus-LLM system detects and explains hate speech across three languages with human-evaluated accuracy that beats LLM-only baselines.
InProceedings of the First Workshop on Trolling, Aggression and Cyberbullying (TRAC- 2018), pages 199–207, Santa Fe, New Mexico, USA
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Explain the Flag: Contextualizing Hate Speech Beyond Censorship
Hybrid vocabulary-plus-LLM system detects and explains hate speech across three languages with human-evaluated accuracy that beats LLM-only baselines.