Automated decision-making systems, especially those based on natural language processing, are pervasive in our lives. They are not only behind the internet search engines we use daily, but also take more critical roles: selecting candidates for a job, determining suspects of a crime, diagnosing autism and more. Such automated systems make errors, which may be harmful in many ways, be it because of the severity of the consequences (as in health issues) or because of the sheer number of people they affect. When errors made by an automated system affect a population more than others, we call the system \textit{biased}.
Most modern natural language technologies are based on artifacts obtained from enormous volumes of text using machine learning, namely language models and word embeddings. Since they are created by applying subsymbolic machine learning, mostly artificial neural networks, they are opaque and practically uninterpretable by direct inspection, thus making it very difficult to audit them.
In this paper, we present a methodology that spells out how social scientists, domain experts, and machine learning experts can collaboratively explore biases and harmful stereotypes in word embeddings and large language models. Our methodology is based on the following principles:
* focus on the linguistic manifestations of discrimination on word embeddings and language models, not on the mathematical properties of the models * reduce the technical barrier for discrimination experts%, be it social scientists, domain experts or other * characterize through a qualitative exploratory process in addition to a metric-based approach * address mitigation as part of the training process, not as an afterthought
Advancing Equitable AI: Evaluating Cultural Expressiveness in LLMs for Latin American Contexts
Department of Applied Computing, Michigan Technological University, Houghton, MI, USA(应用计算系,密歇根技术大学,霍顿,密歇根州,美国);Department of Psychology and Human Factors, Michigan Technological University, Houghton, MI, USA(心理学与人因学系,密歇根技术大学,霍顿,密歇根州,美国)
Towards culturally-appropriate conversational AI for health in the majority world: An exploratory study with citizens and professionals in Latin America
Imperial College London(帝国理工学院);University of Cambridge(剑桥大学);La Victoria Lab(拉维多利亚实验室);Universidad Nacional de Córdoba(科尔多瓦国立大学)
La Leaderboard: A Large Language Model Leaderboard for Spanish Varieties and Languages of Spain and Latin America
SomosNLP;ETSIT, Universidad Politécnica de Madrid(ETSIT,西班牙马德里理工大学);Barcelona Supercomputing Center(巴塞罗那超级计算中心);Hugging Face;Universidad Carlos III de Madrid(马德里卡洛斯三世大学);Instituto de Ingeniería del Conocimiento(知识工程研究所);Centro HiTZ - Ixa, Universidad del País Vasco UPV/EHU(HiTZ-Ixa中心,巴斯克大学);LIACS, Leiden University(LIACS,莱顿大学);Universidad de Jaén(杰嫩大学);Universidad Nacional de Córdoba(科尔多瓦国立大学);Universidad Nacional Autónoma de México(墨西哥国立自治大学);Universidad de la República, Uruguay(乌拉圭共和国大学)
La Leaderboard是一个开源项目,旨在评估西班牙及拉丁美洲各种语言和方言的生成式大型语言模型,通过社区驱动的方式促进多样化语言模型的发展。
Can Small-Scale Data Poisoning Exacerbate Dialect-Linked Biases in Large Language Models?
Department of Electrical and Computer Engineering, Maroun Semaan Faculty of Engineering and Architecture(电气与计算机工程系,马鲁恩·塞马安工程与建筑学院);American University of Beirut(贝鲁特美国大学)
American == White in Multimodal Language-and-Image AI
University of Washington(华盛顿大学)
三个最先进的语言和图像AI模型,CLIP,SLIP和BLIP,评估了以前在社会和实验心理学中观察到的偏见的证据:将美国身份等同于白人。嵌入关联测试(EAT)使用标准化图像的自我识别的亚洲,黑人,拉丁裔/o,和白人个体从芝加哥人脸数据库(CFD)揭示,白人个体更多地与集体内组的话比亚洲,黑人,或拉丁裔/o的个人。在对社会心理学家报告的美国身份的三个核心方面的评估中,单一类别的EAT显示,白人个体的形象与爱国主义和出生在美国有关,但与心理学的先前发现一致,白人个体与不太可能平等对待所有种族和背景的人有关。三个下游机器学习任务显示了将美国人与白人联系起来的偏见。在使用BLIP的视觉问答任务中,97%的白人个体被识别为美国人,而亚洲个体只有3%。当被问及所描绘的个人生活在什么州时,该模型对亚洲人的回答是53%的中国,但对白人的回答总是美国州。在图像描述任务中,BLIP对亚裔个体的种族评论高达36%,但从不对白人个体的种族评论。最后,提供来自CFD的初始化图像和文本“an American person”,使用CLIP的基于文本的指导的合成图像生成器(VQGAN)使所有种族的个体的肤色变亮(基于像素亮度,对于黑人个体,变亮35%)。结果表明,将美国身份等同于白人的偏见是由语言和图像AI学习的,并传播到此类模型的下游应用程序。
Data Caricatures: On the Representation of African American Language in Pretraining Corpora
Columbia University, Department of Computer Science(哥伦比亚大学计算机科学系);University of Michigan, Department of Linguistics(密歇根大学语言学系);University of Pennsylvania, School of Social Policy and Practice, Annenberg School for Communications(宾夕法尼亚大学社会政策与实践学院、安纳伯格通讯学院);University of Tennessee, Knoxville, Department of English(田纳西大学基洛纳分校英语系)
Is It Bad to Work All the Time? Cross-Cultural Evaluation of Social Norm Biases in GPT-4
Department of Computer Science, University of British Columbia(不列颠哥伦比亚大学计算机科学系);Vector Institute for AI(人工智能矢量研究所);Department of Linguistics, University of British Columbia(不列颠哥伦比亚大学语言学系);Mila – Quebec AI Institute(魁北克人工智能研究所);School of Computer Science, McGill University(麦吉尔大学计算机科学学院);Department of Sociology, University of British Columbia(不列颠哥伦比亚大学社会学系)
Kreyòl-MT: Building MT for Latin American, Caribbean and Colonial African Creole Languages
Johns Hopkins University(约翰斯·霍普金斯大学);National Institute of Information and Communications Technology(信息通信研究机构);Brigham Young University(杨百翰大学);Inria Paris(法国国家信息与自动化研究所巴黎分部);Nile University of Nigeria(尼日利亚尼罗河大学);Université Paris Nanterre(巴黎第十大学);University of Minnesota - Twin Cities(明尼苏达大学双城分校);University of Toronto(多伦多大学)