논문해외
How I Aligned Your Model: Scaling Up Multilingual Safety Training with Templates
HF Daily Papers
Safety alignment for Large Language Models is heavily focused on English with limited multilingual representation, particularly damaging low-resource language performance.
원문의 저작권은 HF Daily Papers에 있습니다. 위 발췌는 원문 확인을 돕기 위한 짧은 인용입니다.