中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
Representative Demonstration Selection for In-Context Learning with Two-Stage Determinantal Point Process

文献类型:会议论文

作者Zhao Yang2,3; Yuanzhe Zhang2,3; Dianbo Sui5; Cao Liu1; Jun Zhao2,3; Kang Liu2,3,4
出版日期2023
会议日期2023-12
会议地点Singapore
国家Singapore
英文摘要

Although In-Context Learning has proven effective across a broad array of tasks, its efficiency is noticeably influenced by the selection of demonstrations. Existing methods tend to select different demonstrations for each test instance, which is time-consuming and poses limitations in practical scenarios. Therefore, this study aims to address the challenge of selecting a representative subset of in-context demonstrations that can effectively prompt different test instances in a specific task. We propose that this representative subset should be of high quality and diversity. Our empirical analyses confirm that demonstrations that meet these criteria can indeed bolster model performance. To satisfy these criteria, this paper further introduces a two-stage Determinantal Point Process (DPP) method designed to incorporate both quality and diversity in the process of demonstration selection, thereby obtaining representative in-context demonstrations. Through comprehensive experimentation, we have confirmed the efficacy of our proposed method, paving the way for more practical and effective In-Context Learning.

源URL[http://ir.ia.ac.cn/handle/173211/56724]  
专题复杂系统认知与决策实验室
通讯作者Kang Liu
作者单位1.Meituan, Beijing, China
2.School of Artificial Intelligence, University of Chinese Academy of Sciences, China
3.The Laboratory of Cognition and Decision Intelligence for Complex Systems, Institute of Automation, Chinese Academy of Sciences, China
4.Shanghai Artificial Intelligence Laboratory, China
5.Harbin Institute of Technology, Weihai, China
推荐引用方式
GB/T 7714
Zhao Yang,Yuanzhe Zhang,Dianbo Sui,et al. Representative Demonstration Selection for In-Context Learning with Two-Stage Determinantal Point Process[C]. 见:. Singapore. 2023-12.

入库方式: OAI收割

来源:自动化研究所

浏览0
下载0
收藏0
其他版本

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。