Unmasking camouflage: exploring the challenges of large language models in deciphering African American language & online performativity

Shana Kleiner, Jessica A. Grieser, Shug Miller, James Shepard, Javier Garcia-Perez, Nick Deas, Desmond U. Patton, Elsbeth Turcan, Kathleen McKeown
{"title":"Unmasking camouflage: exploring the challenges of large language models in deciphering African American language & online performativity","authors":"Shana Kleiner,&nbsp;Jessica A. Grieser,&nbsp;Shug Miller,&nbsp;James Shepard,&nbsp;Javier Garcia-Perez,&nbsp;Nick Deas,&nbsp;Desmond U. Patton,&nbsp;Elsbeth Turcan,&nbsp;Kathleen McKeown","doi":"10.1007/s43681-024-00623-2","DOIUrl":null,"url":null,"abstract":"<div><p>The growing accessibility of large language models (LLMs) has raised many questions about the reliability of probabilistically generated natural language responses. While researchers have documented how bias in the training data leads to biased and ethically problematic output, little attention has been paid to the problems which arise from the nature of the varieties of language on which these models are trained. In particular, certain kinds of expressive and performative language use are more common among African American social media users than they occur in the naturalistic speech of African Americans, a discrepancy which models may fail to take into account when they are training on easily-scraped data as being representative of African American speech. Because LLM training data is generally proprietary, in this work we simulate the training data using a collected dataset consisting of 274 posts from Twitter, Reddit, and Hip-Hop lyrics and analyze how LLMs interpreted their meaning. We highlight the difficulties LLMs, including GPT-3 and GPT-4, have in understanding performative AAL and examine how camouflaging and performativity are addressed (or not) by LLMs and demonstrate the harmful implications of misinterpreting online performance.</p></div>","PeriodicalId":72137,"journal":{"name":"AI and ethics","volume":"5 1","pages":"29 - 37"},"PeriodicalIF":0.0000,"publicationDate":"2024-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s43681-024-00623-2.pdf","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"AI and ethics","FirstCategoryId":"1085","ListUrlMain":"https://link.springer.com/article/10.1007/s43681-024-00623-2","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

The growing accessibility of large language models (LLMs) has raised many questions about the reliability of probabilistically generated natural language responses. While researchers have documented how bias in the training data leads to biased and ethically problematic output, little attention has been paid to the problems which arise from the nature of the varieties of language on which these models are trained. In particular, certain kinds of expressive and performative language use are more common among African American social media users than they occur in the naturalistic speech of African Americans, a discrepancy which models may fail to take into account when they are training on easily-scraped data as being representative of African American speech. Because LLM training data is generally proprietary, in this work we simulate the training data using a collected dataset consisting of 274 posts from Twitter, Reddit, and Hip-Hop lyrics and analyze how LLMs interpreted their meaning. We highlight the difficulties LLMs, including GPT-3 and GPT-4, have in understanding performative AAL and examine how camouflaging and performativity are addressed (or not) by LLMs and demonstrate the harmful implications of misinterpreting online performance.

求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信