当前位置: X-MOL 学术Communication Methods and Measures › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
On Measurement Validity and Language Models: Increasing Validity and Decreasing Bias with Instructions
Communication Methods and Measures ( IF 6.3 ) Pub Date : 2024-07-29 , DOI: 10.1080/19312458.2024.2378690
Moritz Laurer 1 , Wouter van Atteveldt 1 , Andreu Casas 2 , Kasper Welbers 1
Affiliation  

Language models like BERT or GPT are becoming increasingly popular measurement tools, but are the measurements they produce valid? Literature suggests that there is still a relevant gap between the...

中文翻译:


关于测量有效性和语言模型:通过指令提高有效性并减少偏差



像 BERT 或 GPT 这样的语言模型正在成为越来越流行的测量工具,但它们产生的测量有效吗?文献表明,两者之间仍然存在相关差距……
更新日期:2024-08-03
down
wechat
bug