{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/do-rnns-learn-human-like-abstract-word-order","title":"Do RNNs learn human-like abstract word order preferences?","arxiv_id":"1811.01866","date":"2018-11-05","proceeding":"WS 2019 1","authors":["Richard Futrell","Roger P. Levy"],"abstract":"RNN language models have achieved state-of-the-art results on various tasks,\nbut what exactly they are representing about syntax is as yet unclear. Here we\ninvestigate whether RNN language models learn humanlike word order preferences\nin syntactic alternations. We collect language model surprisal scores for\ncontrolled sentence stimuli exhibiting major syntactic alternations in English:\nheavy NP shift, particle shift, the dative alternation, and the genitive\nalternation. We show that RNN language models reproduce human preferences in\nthese alternations based on NP length, animacy, and definiteness. We collect\nhuman acceptability ratings for our stimuli, in the first acceptability\njudgment experiment directly manipulating the predictors of syntactic\nalternations. We show that the RNNs' performance is similar to the human\nacceptability ratings and is not matched by an n-gram baseline model. Our\nresults show that RNNs learn the abstract features of weight, animacy, and\ndefiniteness which underlie soft constraints on syntactic alternations.","url_abs":"http://arxiv.org/abs/1811.01866v1","url_pdf":"http://arxiv.org/pdf/1811.01866v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"do-rnns-learn-human-like-abstract-word-order","repo_url":"https://github.com/langprocgroup/rnn_soft_constraints","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"sentence","task_name":"Sentence"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1811.01866","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}