HatePrototypes built from few examples enable cross-benchmark transfer and efficient detection for both explicit and implicit hate speech.
These prototypes are further used for cross-task trans- fer analysis and layer-wise classification for model early exiting
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
HatePrototypes: Interpretable and Transferable Representations for Implicit and Explicit Hate Speech Detection
HatePrototypes built from few examples enable cross-benchmark transfer and efficient detection for both explicit and implicit hate speech.