A new 3,050-instance benchmark shows that injecting linguistic knowledge graphs into prompts raises LLM borrowing-classification accuracy from 25-35% to 71-81% for Luxembourgish while neology detection remains difficult.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Do LLMs Know What Luxembourgish Borrows? Probing Lexical Neology in Low-Resource Multilingual Models
A new 3,050-instance benchmark shows that injecting linguistic knowledge graphs into prompts raises LLM borrowing-classification accuracy from 25-35% to 71-81% for Luxembourgish while neology detection remains difficult.