Earticle

초록

영어

This study analyzes cultural biases in major large language models from the United States, South Korea, and China (GPT-4, CLOVA X, and Qwen1.5) through story generation tasks using culture-specific names. Morphological analysis of the generated stories revealed that all models exhibited certain cultural biases. GPT-4 did not show negative biases toward Korean and Chinese cultures but tended to prefer traditional and rural settings when describing these cultures. In contrast, CLOVA X and Qwen1.5, which are specialized for their respective national languages, portrayed their own cultures in modern and positive terms while using a relatively higher proportion of negative adjectives and unrealistic settings when describing Western contexts. These findings are significant because they go beyond the conventional focus on biases in Western-centric models toward non-Western contexts. They newly reveal that East Asian-based models can also exhibit similar biases when representing Western cultures. This research suggests that current language model has fundamental limitations in achieving cultural neutrality and highlights the importance of balanced learning and reflection of diverse cultural contexts as a crucial challenge in language model development.

목차

ABSTRACT
1. 서론
2. 이론적 배경
3. 실험 방법론
3.1. 데이터
3.2. 모델
3.3. 프롬프트
3.4. 생성 과정
4. 실험 결과
4.1. 형용사 분석
4.2. 일반명사 분석
5. 결론
참고문헌
부록

키워드

저자

• 정가연 [ 연세대학교 언어정보연구원 / 연구원 ]
• 강채안 [ 연세대학교 언어정보연구원 / 연구원 ]
• 김민선 [ 연세대학교 / 대학원생 ]
• 노소현 [ 연세대학교 / 대학원생 ]
• 최혜지 [ 연세대학교 / 대학원생 ]
• 김한샘 [ 연세대학교 / 교수 ]

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  Corpus Linguistics Research

• 간기

  계간

• pISSN

  2465-812X

• 수록기간

  2015~2025

• 십진분류

  KDC 701 DDC 410