LLMic, a 3B Romanian-English model with a diacritic-free tokenizer, reports higher WMT English-to-Romanian translation scores than larger open models, though the evaluation setup is biased.
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
-
LLMic: Romanian Foundation Language Model
LLMic, a 3B Romanian-English model with a diacritic-free tokenizer, reports higher WMT English-to-Romanian translation scores than larger open models, though the evaluation setup is biased.