BriefGPT.xyz
Jun, 2023
自然语言处理中社会人口统计信息偏差调查
Survey on Sociodemographic Bias in Natural Language Processing
HTML
PDF
Vipul Gupta, Pranav Narayanan Venkit, Shomir Wilson, Rebecca J. Passonneau
TL;DR
本篇论文调查了209篇关于自然语言处理模型中的偏差的论文，提出了社会人口统计学偏见的定义，并确定了研究偏见的三个主要类别：偏见类型，偏见度量和去偏见。作者总结说，目前的去偏见技术是肤浅的，不能真正消除偏见；最后提供了未来工作的建议。
Abstract
deep neural networks
often learn
unintended biases
during training, which might have harmful effects when deployed in real-world settings. This paper surveys 209 papers on bias in
→