[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-82544-en":3,"doc-seo-82544-105":29,"detail-sidebar-cat-0-en-105":83},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},82544,549758146520,"Patrick","https://ap-avatar.wpscdn.com/avatar/80002397d8c0411e94?_k=1775819394049821470",8,"Research & Report","Multi-Label Node Classification with Label Influence Propagation","Multi-label node classification (MLNC) on graphs is central in domains such as protein function prediction and user preference modeling. Existing methods exploit label co-occurrence or label proximity via graph neural networks and label embeddings, yet they do not model complex, non-Euclidean label interactions within the message-passing process. The proposed approach decomposes GNN message passing into propagation and transformation, quantifies label influence correlations, builds a label influence graph, and propagates high-order influences to amplify beneficial labels and mitigate harmful ones, achieving consistent SOTA gains on benchmark datasets.","arXiv :2607 .0067 1v 1 [ cs .LG] 1 Jul 2026  \nMULTI-LABEL NODE CLASSIFICATION WITH LABEL INFLUENCE PROPAGATION  \nYifei Sun 1 , Zemin Liu 1 ∗, Bryan Hooi2 , Yang Yang 1 ∗, Rizal Fathony3 , Jia Chen4 , Bingsheng He2 ∗  \n1Zhejiang University, 2National University of Singapore,  \n3 Capital One, 4 GrabTaxi Holdings Pte. Ltd.  \n{yifeisun, liu.zemin, [yangya](yangya}@zju.edu.cn)[}](yangya}@zju.edu.cn)[@zju.edu.cn](yangya}@zju.edu.cn) ,{bhooi, [hebs](hebs}@comp.nus.edu.sg)[}](hebs}@comp.nus.edu.sg)[@comp.nus.edu.sg](hebs}@comp.nus.edu.sg) ,  \n[rizal.fathony@capitalone.com](rizal.fathony@capitalone.com) , [jia.chen@grab.com](jia.chen@grab.com)  \nABSTRACT  \nGraphs are a complex and versatile data structure used across various domains, with possibly multi-label nodes playing a particularly crucial role. Examples include proteins in PPI networks with multiple functions and users in social ore-commerce networks exhibiting diverse interests. Tackling multi-label node classification (MLNC) on graphs has led to the development of various approaches. Some methods leverage graph neural networks (GNNs) to exploit label co-occurrence correlations, while others incorporate label embeddings to capture label proximity. However, these approaches fail to account for the intricate influences between labels in non-Euclidean graph data. To address this issue, we decompose the message passing process in GNNs into two operations: propagation and transformation. We then conduct a comprehensive analysis and quantification of the influence correlations between labels in each operation. Building on these insights, we propose a novel model, Label Influence Propagation (LIP) . Specifically, we construct a label influence graph based on the integrated label correlations. Then, we propagate high-order influences through this graph, dynamically adjusting the learning process by amplifying labels with positive contributions and mitigating those with negative influence. Finally, our framework is evaluated on comprehensive benchmark datasets, consistently outperforming SOTA methods across various settings, demonstrating its effectiveness on MLNC tasks 1.  \n1 INTRODUCTION  \nGraphs, as a complex data structure, are prevalent across various fields (Jiang et al., 2019; Kipf & Welling, 2016; Ying et al., 2018; Liu et al., 2023; Fang et al., 2024) . Among these, graphs with multi-label nodes are common and of great importance. For instance, proteins in ogbn-protein dataset have multiple functions (Hu et al., 2020) . Accurately identifying all the functions can assist with understanding biological processes and advancing biomedical research. Thus, we focus on this realistic but challenging problem named multi-label node classification on graphs, which we abbreviate as MLNC in the following paper.  \nPrior studies. Current methods typically adopt three strategies to address this problem. The first strategy is to neglect the multi-label information and predict the labels without mining label correlations (Shi et al., 2020b; Li et al., 2023) . The second strategy is to explicitly treat labels as anew type of node and incorporate them into the original graph, thereby enhancing task performance through propagation and aggregation information between nodes and label nodes (Gao et al., 2019; Shi et al., 2020a) . Since only incomplete connections between nodes and label nodes are available in the training set, the third strategy is to integrate label representations into the neighbor aggregation and classification processes, thereby improving the utilization of multi-label information (Zhou et al., 2021; Xiao et al., 2022) . However, these strategies underestimate the complex label correla-  \n∗ Corresponding authors.  \n1Our code is available at [https://github.com/Xtra-Computing/LIP_MLNC](https://github.com/Xtra-Computing/LIP_MLNC).  \n3 2 1 0  \n(a) AP Difference on DBLP  \n0 1 2 3  \n2  \n1  \n0  \n1  \n2  \n(b) AP Difference on BlogCat  \n9 8 7 6 5 4 3 2 1 0  \n0 1 2 3 4 5 6 7 8 9  \n0.2  \n0.","cbCaifRvfDqfgcYo","https://ap.wps.com/l/cbCaifRvfDqfgcYo","pdf",1044670,1,25,"English","en",105,"# Abstract\n# Introduction\n## Prior studies\n## Observations\n## Challenges","[{\"question\":\"How is the effectiveness of the model evaluated?\",\"answer\":\"The framework is evaluated on comprehensive benchmark datasets and shows consistent improvements over state-of-the-art methods across different settings.\"}]",1784181450,63,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":78,"head_meta":80,"extra_data":82,"updated_unix":27},"multi-label-node-classification-with-label-influence-propagation","",{"@graph":35,"@context":77},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/multi-label-node-classification-with-label-influence-propagation/82544/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71],{"name":72,"@type":73,"acceptedAnswer":74},"How is the effectiveness of the model evaluated?","Question",{"text":75,"@type":76},"The framework is evaluated on comprehensive benchmark datasets and shows consistent improvements over state-of-the-art methods across different settings.","Answer","https://schema.org",{"og:url":51,"og:type":79,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":81,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":84},[85,89,93,97,102,107,112,115,120,123,127],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":86,"show_sort_weight":87,"slug":88},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":90,"show_sort_weight":91,"slug":92},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Exam",70,"exam",{"id":98,"doc_module":4,"doc_module_name":45,"category_name":99,"show_sort_weight":100,"slug":101},5,"Comic",60,"comic",{"id":103,"doc_module":4,"doc_module_name":45,"category_name":104,"show_sort_weight":105,"slug":106},6,"Technology",50,"technology",{"id":108,"doc_module":4,"doc_module_name":45,"category_name":109,"show_sort_weight":110,"slug":111},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":113,"slug":114},30,"research-report",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":45,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":124,"slug":126},10,"Lifestyle","lifestyle",{"id":128,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":98,"slug":130},19,"General","general"]