Abstract
In AI, most evaluations of natural language understanding tasks are conducted in standardized dialects such as Standard American English (SAE). In this work, we investigate how accurately large language models (LLMs) represent African American Vernacular English (AAVE). We analyze three LLMs to compare their usage of AAVE to the usage of humans who natively speak AAVE. We first analyzed interviews from the Corpus of Regional African American Language and TwitterAAE to identify the typical contexts where people use AAVE grammatical features such as ain't. We then prompted the LLMs to produce text in AAVE and compared the model-generated text to human usage patterns. We find that, in many cases, there are substantial differences between AAVE usage in LLMs and humans: LLMs usually underuse and misuse grammatical features characteristic of AAVE. Furthermore, through sentiment analysis and manual inspection, we found that the models replicated stereotypes about African Americans. These results highlight the need for more diversity in training data and the incorporation of fairness methods to mitigate the perpetuation of stereotypes.
Abstract (translated)
在人工智能领域,大多数关于自然语言理解任务的评估都是基于标准方言进行的,例如标准美式英语(SAE)。在这项工作中,我们研究了大规模语言模型(LLMs)对非裔美国人英语变体(AAVE)的准确表示程度。我们分析了三种不同的大型语言模型,并将其使用AAVE的情况与母语为AAVE的人类使用者进行比较。首先,我们通过《区域非裔美国人语言语料库》和TwitterAAE中的访谈来识别人们在何种典型情境下使用如“ain't”这样的AAVE语法特征。然后,我们提示这些LLMs生成具有AAVE特点的文本,并将其与人类的实际使用模式进行了对比。我们发现,在许多情况下,LLMs中使用的AAVE与人类使用者之间存在显著差异:大型语言模型通常会低估或错误使用典型的AAVE语法特征。 此外,通过情感分析和人工检查,我们还发现这些模型复制了关于非裔美国人的刻板印象。这些结果强调了在训练数据中需要更多的多样性,并且必须纳入公平性方法以减少刻板印象的延续。
URL
https://arxiv.org/abs/2602.21485