作者位次 #3/5ACL 2026Poster· Natural Language Processing
EIFFEL : a novel benchmark to measure bias of E nglish heavy training on F rench idiomatic expressions
Charlotte Noel · Nicholas Asher · Olivier Gouvert · Farah Benamara · Julie Hunter
摘要
Mainstream multilingual LLMs are generally trained on a much higher proportion of English than multilingual data, raising questions about their ability to capture linguistic features particular to non-English languages or to capture information important to non-anglophone cultures. We add to a gr...