상세 보기
초록
Recent natural language processing technology has been advancing at an unprecedented pace, driven by the development of large language models. However, the issue of hallucination, where the model generates inaccurate or nonsensical responses, remains a challenge to be addressed. This paper analyzes various prompt engineering techniques in large-scale language models and derives prompt engineering methods that can achieve optimal response performance for each dataset. The study found that the most suitable prompt engineering techniques can vary depending on the characteristics of each dataset. © 2025, Korean Institute of Communications and Information Sciences. All rights reserved.
키워드
Chain of Thought; In-context learning; Large Language Model; Prompt Engineering; Retrieval-Augmented Generation
- 제목
- Research on prompt engineering techniques in large language models
- 제목 (타언어)
- Research on Prompt Engineering Techniques in Large Language Models
- 저자
- Son, Minjun; Lee, Sungjin
- 발행일
- 2025-01
- 유형
- Article
- 저널명
- 한국통신학회논문지
- 권
- 50
- 호
- 1
- 페이지
- 9 ~ 21