论文标题
探测人类默示假设的神经语言模型
Probing Neural Language Models for Human Tacit Assumptions
论文作者
论文摘要
人类具有刻板印象的默契假设（Stas）（Prince，1978）或对通用概念的命题信念。这种关联对于理解自然语言至关重要。我们构建了一组单词预测集的诊断集提示，以评估最近在大型文本语料库捕获Stas训练的神经上下文化语言模型。我们的提示是基于人类对概念关联的心理研究的反应。我们发现，在给定相关特性的情况下，模型在检索概念方面非常有效。我们的结果证明了经验证据表明，刻板印象的概念表示是在半监督语言暴露的神经模型中捕获的。
Humans carry stereotypic tacit assumptions (STAs) (Prince, 1978), or propositional beliefs about generic concepts. Such associations are crucial for understanding natural language. We construct a diagnostic set of word prediction prompts to evaluate whether recent neural contextualized language models trained on large text corpora capture STAs. Our prompts are based on human responses in a psychological study of conceptual associations. We find models to be profoundly effective at retrieving concepts given associated properties. Our results demonstrate empirical evidence that stereotypic conceptual representations are captured in neural models derived from semi-supervised linguistic exposure.
