A trainable pipeline of translation models and a frozen LLM improves Hebrew last-word prediction accuracy to 5.59%, about twice the best baseline.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
TALL -- A Trainable Architecture for Enhancing LLM Performance in Low-Resource Languages
A trainable pipeline of translation models and a frozen LLM improves Hebrew last-word prediction accuracy to 5.59%, about twice the best baseline.