{"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/modeling-semantic-plausibility-by-injecting","title":"Modeling Semantic Plausibility by Injecting World Knowledge","arxiv_id":"1804.00619","date":"2018-04-02","proceeding":"NAACL 2018 6","authors":["Su Wang","Greg Durrett","Katrin Erk"],"abstract":"Distributional data tells us that a man can swallow candy, but not that a man\ncan swallow a paintball, since this is never attested. However both are\nphysically plausible events. This paper introduces the task of semantic\nplausibility: recognizing plausible but possibly novel events. We present a new\ncrowdsourced dataset of semantic plausibility judgments of single events such\nas \"man swallow paintball\". Simple models based on distributional\nrepresentations perform poorly on this task, despite doing well on selection\npreference, but injecting manually elicited knowledge about entity properties\nprovides a substantial performance boost. Our error analysis shows that our new\ndataset is a great testbed for semantic plausibility models: more sophisticated\nknowledge representation and propagation could address many of the remaining\nerrors.","url_abs":"http://arxiv.org/abs/1804.00619v3","url_pdf":"http://arxiv.org/pdf/1804.00619v3.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":"modeling-semantic-plausibility-by-injecting","repo_url":"https://github.com/suwangcompling/Modeling-Semantic-Plausibility-NAACL18","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"world-knowledge","task_name":"World Knowledge"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1804.00619","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}