Three small language models vary in retaining fine-grained emotions during backtranslation, with emotion-aware prompting providing improvement and ModernBERT performing similarly to BERT for classification.
write newline
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
fields
cs.CL 3verdicts
UNVERDICTED 3representative citing papers
Tokenization scheme performance in Arabic-English MT depends on whether statistical or neural models are used and on data size, with hybrid system selection providing gains.
Machine translation systems lose lexical richness relative to human translations and may thereby exacerbate biases such as gender bias.
citing papers explorer
-
Beyond Semantics: Measuring Fine-Grained Emotion Preservation in Small Language Model-Based Machine Translation
Three small language models vary in retaining fine-grained emotions during backtranslation, with emotion-aware prompting providing improvement and ModernBERT performing similarly to BERT for classification.
-
The Impact of Preprocessing on Arabic-English Statistical and Neural Machine Translation
Tokenization scheme performance in Arabic-English MT depends on whether statistical or neural models are used and on data size, with hybrid system selection providing gains.
-
Lost in Translation: Loss and Decay of Linguistic Richness in Machine Translation
Machine translation systems lose lexical richness relative to human translations and may thereby exacerbate biases such as gender bias.