# Probing Neural Language Models for Human Tacit Assumptions

arXiv: https://arxiv.org/abs/2004.04877 · HTML: https://mlsanity.com/paper/2004.04877/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 10 Apr 2020.

## Built on by (most cited first, 1 of 1)

- [Visual Commonsense in Pretrained Unimodal and Multimodal Models](https://mlsanity.com/paper/2205.01850/) (2022)

## Abstract (arXiv)

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.