{"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/community-regularization-of-visually-grounded","title":"Community Regularization of Visually-Grounded Dialog","arxiv_id":"1808.04359","date":"2018-08-10","proceeding":null,"authors":["Akshat Agarwal","Swaminathan Gurumurthy","Vasu  Sharma","Mike Lewis","Katia Sycara"],"abstract":"The task of conducting visually grounded dialog involves learning\ngoal-oriented cooperative dialog between autonomous agents who exchange\ninformation about a scene through several rounds of questions and answers in\nnatural language. We posit that requiring artificial agents to adhere to the\nrules of human language, while also requiring them to maximize information\nexchange through dialog is an ill-posed problem. We observe that humans do not\nstray from a common language because they are social creatures who live in\ncommunities, and have to communicate with many people everyday, so it is far\neasier to stick to a common language even at the cost of some efficiency loss.\nUsing this as inspiration, we propose and evaluate a multi-agent\ncommunity-based dialog framework where each agent interacts with, and learns\nfrom, multiple agents, and show that this community-enforced regularization\nresults in more relevant and coherent dialog (as judged by human evaluators)\nwithout sacrificing task performance (as judged by quantitative metrics).","url_abs":"http://arxiv.org/abs/1808.04359v2","url_pdf":"http://arxiv.org/pdf/1808.04359v2.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":"community-regularization-of-visually-grounded","repo_url":"https://github.com/agakshat/visualdialog-pytorch","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1808.04359","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}