A standard fine-tuned multilingual transformer reaches 88.4% accuracy for hate speech detection and 66.1% for target identification on a five-language Devanagari benchmark.
Nikolaevna Alexeeva, Aleksandr Maximovich Ryabchikov, Calambur Sivaramamurti, and Yury Konstantinovich Yefremov
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
IITR-CIOL@NLU of Devanagari Script Languages 2025: Multilingual Hate Speech Detection and Target Identification in Devanagari-Scripted Languages
A standard fine-tuned multilingual transformer reaches 88.4% accuracy for hate speech detection and 66.1% for target identification on a five-language Devanagari benchmark.