BriefGPT.xyz
Mar, 2024
生成语言模型的非歧视标准
Non-discrimination Criteria for Generative Language Models
HTML
PDF
Sara Sterlie, Nina Weng, Aasa Feragen
TL;DR
研究如何发现和量化生成语言模型中的性别偏见，并设计了针对职业性别刻板印象的标准，通过职业性别刻板印象的测试结果论证了这些标准在生成人工智能模型中的存在。
Abstract
Within recent years,
generative ai
, such as large
language models
, has undergone rapid development. As these models become increasingly available to the public, concerns arise about perpetuating and amplifying ha
→