This project investigates whether modern large language models (LLMs) understand concept variation across languages. We focus on the task of lexical selection—choosing the most suitable lexeme in the ...