{"pk":29408,"title":"Probing Neural Language Models for Human Tacit Assumptions","subtitle":null,"abstract":"Humans carry stereotypic tacit assumptions (STAs) (Prince,1978), or propositional beliefs about generic concepts. Suchassociations are crucial for understanding natural language.We construct a diagnostic set of word prediction prompts toevaluate whether recent neural contextualized language mod-els trained on large text corpora capture STAs. Our promptsare based on human responses in a psychological study of con-ceptual associations. We find models to be profoundly effec-tive at retrieving concepts given associated properties. Our re-sults demonstrate empirical evidence that stereotypic concep-tual representations are captured in neural models derived fromsemi-supervised linguistic exposure.","language":"eng","license":{"name":"","short_name":"","text":null,"url":""},"keywords":[{"word":"language models; deep neural networks; conceptrepresentations; norms; semantics"}],"section":"Neural Networks","is_remote":true,"remote_url":"https://escholarship.org/uc/item/95d5n9s0","frozenauthors":[{"first_name":"Nathaniel","middle_name":"","last_name":"Weir","name_suffix":"","institution":"Johns Hopkins University","department":""},{"first_name":"Adam","middle_name":"","last_name":"Poliak","name_suffix":"","institution":"Johns Hopkins University","department":""},{"first_name":"Benjamin","middle_name":"Van","last_name":"Durme","name_suffix":"","institution":"Johns Hopkins University","department":""}],"date_submitted":null,"date_accepted":null,"date_published":"2020-01-01T18:00:00Z","render_galley":null,"galleys":[{"label":"PDF","type":"pdf","path":"https://journalpub.escholarship.org/cognitivesciencesociety/article/29408/galley/19268/download/"}]}