Probing Neural Language Models for Human Tacit Assumptions
- Nathaniel Weir, Department of Computer Science, Johns Hopkins University, Baltimore, Maryland, United States
- Adam Poliak, Department of Computer Science, Johns Hopkins University, Baltimore, Maryland, United States
- Benjamin Van Durme, Department of Computer Science, Johns Hopkins University, Baltimore, Maryland, United States
AbstractHumans carry stereotypic tacit assumptions (STAs), 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.