{"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/comparatives-quantifiers-proportions-a-multi","title":"Comparatives, Quantifiers, Proportions: A Multi-Task Model for the Learning of Quantities from Vision","arxiv_id":"1804.05018","date":"2018-04-13","proceeding":"NAACL 2018 6","authors":["Sandro Pezzelle","Ionut-Teodor Sorodoc","Raffaella Bernardi"],"abstract":"The present work investigates whether different quantification mechanisms\n(set comparison, vague quantification, and proportional estimation) can be\njointly learned from visual scenes by a multi-task computational model. The\nmotivation is that, in humans, these processes underlie the same cognitive,\nnon-symbolic ability, which allows an automatic estimation and comparison of\nset magnitudes. We show that when information about lower-complexity tasks is\navailable, the higher-level proportional task becomes more accurate than when\nperformed in isolation. Moreover, the multi-task model is able to generalize to\nunseen combinations of target/non-target objects. Consistently with behavioral\nevidence showing the interference of absolute number in the proportional task,\nthe multi-task model no longer works when asked to provide the number of target\nobjects in the scene.","url_abs":"http://arxiv.org/abs/1804.05018v1","url_pdf":"http://arxiv.org/pdf/1804.05018v1.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":"comparatives-quantifiers-proportions-a-multi","repo_url":"https://github.com/sandropezzelle/multitask-quant","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1804.05018","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}