[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-155114-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-155114-en":131},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","confidence-aware-with-prototype-alignment-for-partial-multi-label-learning","Confidence-Aware With Prototype Alignment for Partial Multi-label Learning","","Label prototype learning has become an effective paradigm for Partial Multi-Label Learning by modeling structured label semantics and estimating confidence through prototypes, while filtering noise. Existing prototype-based approaches suffer because class prototypes become biased under noisy candidate labels, especially when positive samples are scarce. The work proposes mutually class prototype alignment using two transformation matrices to correct prototypes via fuzzy clustering and candidate-label alignment. It also leverages a learned permutation matrix to transform fuzzy memberships for label reliability estimation, enabling robust multilabel classifier training. Experiments show clear gains over state-of-the-art PML methods.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/confidence-aware-with-prototype-alignment-for-partial-multi-label-learning/155114/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/confidence-aware-with-prototype-alignment-for-partial-multi-label-learning/155114.png","ImageObject",300,407,{"name":42,"@type":43},"Jacob","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-30","2026-08-28",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",14,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"What problem does the paper address in partial multi-label learning?","Question",{"text":63,"@type":64},"It addresses label disambiguation from uncertain candidate label sets that contain both true labels and noisy false positive labels, especially under complex noise patterns and scarce positive samples.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"How does the proposed method improve class prototype learning?",{"text":68,"@type":64},"It introduces mutually class prototype alignment using two different transformation matrices, so prototypes learned by fuzzy clustering and candidate label sets align and correct each other despite noise.",{"name":70,"@type":61,"acceptedAnswer":71},"How does the method handle noise during classifier learning?",{"text":72,"@type":64},"It learns a permutation matrix to transform fuzzy memberships into a label reliability indicator matrix, reducing numerical values in non-label regions and limiting the introduction of incorrect labels.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},155114,1787905383,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,111,115,120,123,127],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":112,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":113,"slug":114},8,30,"research-report",{"id":116,"doc_module":4,"doc_module_name":25,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":25,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":124,"doc_module":4,"doc_module_name":25,"category_name":125,"show_sort_weight":124,"slug":126},10,"Lifestyle","lifestyle",{"id":128,"doc_module":4,"doc_module_name":25,"category_name":129,"show_sort_weight":97,"slug":130},19,"General","general",{"code":4,"msg":82,"data":132},{"doc_id":79,"user_id":133,"nickname":42,"user_avatar":134,"doc_module":4,"category_id":112,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":118,"language":140,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":12,"update_tm":80,"read_time":104,"preview_page_count":4},962084931830,"https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd","Confidence-Aware With Prototype Alignment for Partial Multi-label Learning  \nWeijun Lv 1 , Yu Chen 1 , Xiaozhao Fang 1∗ , Xuhuan Zhu 1 , Jie Wen2 ∗, Guoxu Zhou 1 , Sixian Chan3  \n1 School of Automation, Guangdong University of Technology, Guangzhou 510006, China  \n2 Shenzhen Key Laboratory of Visual Object Detection and Recognition,  \nHarbin Institute of Technology (Shenzhen), Shenzhen 518055, China  \n3 College of Computer Science and Technology, Zhejiang University of Technology,  \nHangzhou 310023, China  \n{lvweijun0201, [chenyu9265324}@163.com](chenyu9265324}@163.com) , {xzhfang168, [xuhuanz_zz}@126.com](xuhuanz_zz}@126.com)  \n[jiewen_pr@126.com](jiewen_pr@126.com) , [gx.zhou@gdut.edu.cn](gx.zhou@gdut.edu.cn) , [sxchan@zjut.edu.cn](sxchan@zjut.edu.cn)  \nAbstract  \nLabel prototype learning has emerged as an effective paradigm in Partial MultiLabel Learning (PML), providing a distinctive framework for modeling structured representations of label semantics while naturally filtering noise through prototypebased label confidence estimation. However, existing prototype-based methods face a critical limitation: class prototypes are the biased estimates due to noisy candidate labels, particularly when positive samples are scarce. To this end, we first propose a mutually class prototype alignment strategy bypassing noise interference by introducing two different transformation matrices, which makes the class prototypes learned by the fuzzy clustering and candidate label set mutually alignment for correcting themselves. Such alignment is also passed on to the fuzzy memberships label in turn. In addition, to eliminate noise interference in the candidate label set during the classifier learning, we use the learned permutation matrix to transform the fuzzy memberships label for learning a label reliability indicator matrix accompanied by the candidate label set. This makes the label reliability indicator matrix absolutely prevent the occurrence of numerical values located in non-label and simultaneously eliminate the introduction of incorrect label as much as possible. The resulting indicator matrix guides a robust multilabel classifier training process, jointly optimizing label confidence and classifier parameters. Extensive experiments demonstrate that our proposed model exhibits significant performance advantages over state-of-the-art PML approaches.  \n1 Introduction  \nMulti-Label Learning (MLL) is an important branch of machine learning that allows an instance to belong to multiple categories simultaneously, with widespread applications in image annotation[1, 2], text categorization[3], and medical diagnosis [4] . However, in practical application scenarios, obtaining precisely annotated data is often difficult due to high annotation costs and inherent ambiguity[5, 6, 7, 8] . Annotators frequently provide a candidate label set when uncertain, containing both ground-truth labels and noisy labels (i.e., false positive labels mistakenly included) . For example, in image annotation tasks, objects with similar visual appearances may be incorrectly labeled, which will introduce interference to the classification prediction. To address this issue, Partial Multi-Label Learning (PML) [9] has emerged as a new weakly supervised learning framework. The  \n∗ Corresponding Author  \n39th Conference on Neural Information Processing Systems (NeurIPS 2025) .  \nmain task of PML is to learn from this uncertain supervisory information to accurately predict the ground-truth labels for unknown instances.  \nPML research primarily focuses on identifying ground-truth labels from the candidate label set, known as label disambiguation. Existing methods mostly adopt explicit learning strategies, directly removing noisy labels from candidate label sets through various techniques. For instance, some methods use low-rank and sparse decomposition to separate ground-truth labels from noisy ones [10, 11]; others estimate the credibility of candidate labels through labe","cbCaijKaIT9JR18W","https://ap.wps.com/l/cbCaijKaIT9JR18W","pdf",2663082,"English","# Introduction\n## Multi-Label Learning and Candidate Labels\n## Label Disambiguation in Partial Multi-Label Learning\n## Limitations of Existing Approaches\n## Prototype-Based Learning Motivation","[{\"question\":\"What problem does the paper address in partial multi-label learning?\",\"answer\":\"It addresses label disambiguation from uncertain candidate label sets that contain both true labels and noisy false positive labels, especially under complex noise patterns and scarce positive samples.\"},{\"question\":\"How does the proposed method improve class prototype learning?\",\"answer\":\"It introduces mutually class prototype alignment using two different transformation matrices, so prototypes learned by fuzzy clustering and candidate label sets align and correct each other despite noise.\"},{\"question\":\"How does the method handle noise during classifier learning?\",\"answer\":\"It learns a permutation matrix to transform fuzzy memberships into a label reliability indicator matrix, reducing numerical values in non-label regions and limiting the introduction of incorrect labels.\"}]","Confidence-Aware With Prototype Alignment for Partial Multi-label Learning | PDF"]