{"id":15015,"date":"2024-07-08T08:55:43","date_gmt":"2024-07-08T00:55:43","guid":{"rendered":"https:\/\/www.1ai.net\/?p=15015"},"modified":"2024-07-08T08:55:43","modified_gmt":"2024-07-08T00:55:43","slug":"%e6%96%b0%e9%a2%96%e5%a4%9a%e6%a8%a1%e6%80%81%e6%8e%a8%e8%8d%90%e7%b3%bb%e7%bb%9f%e8%8c%83%e5%bc%8fdiffmm%ef%bc%8c%e8%ae%a9%e6%89%a9%e6%95%a3%e6%a8%a1%e5%9e%8b%e4%b9%9f%e8%83%bd%e6%8e%a8%e8%8d%90","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/15015.html","title":{"rendered":"The novel multimodal recommendation system paradigm DiffMM allows the diffusion model to recommend short videos!"},"content":{"rendered":"<p>Researchers from HKU and Tencent have proposed a new paradigm for multimodal recommender systems -- the<a href=\"https:\/\/www.1ai.net\/en\/tag\/diffmm\" title=\"_Other Organiser\" target=\"_blank\" >DiffMM<\/a>The aim is to increase<a href=\"https:\/\/www.1ai.net\/en\/tag\/%e7%9f%ad%e8%a7%86%e9%a2%91\" title=\"[View articles tagged with [short video]]\" target=\"_blank\" >Short Video<\/a>Recommendation accuracy. The system achieves more accurate recommendations by creating a graph containing information about users and videos and utilizing graph diffusion and contrast learning techniques to better understand the relationship between users and videos.<\/p>\n<p>DiffMM ' s model methodology consists of three main components: multi-mode map diffusion model, multi-mode map aggregation and cross-mode comparison enhancement. Among them, the multi-modular diffusion model uses a model to detect the probability of noise diffusion, aligning the user-matter synergetic signal with the multi-modular information and effectively addressing the negative effects of the multi-modular referral system. At the same time, the production and optimization of model sensory images has been achieved through the optimization of the diffusion of the probabilistic proliferation paradigm and model perception\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-15016\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2024\/07\/6385595356882449226660883.png\" alt=\"\" width=\"629\" height=\"143\" \/><\/p>\n<p>In terms of cross-modal contrast enhancement, DiffMM utilizes modality-aware contrast view and contrast enhancement methods to capture the consistency of user interaction patterns on different item modalities and improve recommender system performance.<\/p>\n<p>Paper:https:\/\/arxiv.org\/abs\/2406.1178<\/p>","protected":false},"excerpt":{"rendered":"<p>A new model of a multi-state referral system - DiffMM - has been proposed by Hong Kong ' s research staff to improve the accuracy of short video referrals. The system achieves more accurate recommendations by creating a map of users and video information and using the diffusion and comparative learning techniques to better understand the relationship between users and videos. DiffMM ' s model methodology consists of three main components: multi-mode map diffusion model, multi-mode map aggregation and cross-mode comparison enhancement. Among them, the multi-modular diffusion model uses a model to detect the probability of noise diffusion, aligning the user-matter synergetic signal with the multi-modular information and effectively addressing the negative effects of the multi-modular referral system. At the same time, modeling is achieved by optimizing the diffusion of the probabilistic proliferation paradigm and model perception<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[3389,1007],"collection":[],"class_list":["post-15015","post","type-post","status-publish","format-standard","hentry","category-news","tag-diffmm","tag-1007"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/15015","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/comments?post=15015"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/15015\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=15015"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=15015"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=15015"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=15015"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}