[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_291":292},"routes/_layout",{"_11":211},"routes/entity.$type.$id",{"_11":12,"_13":14,"_15":16,"_17":18,"_19":20,"_21":22,"_23":24,"_25":26,"_27":28},"dehydratedState",{"_73":74,"_75":76},"type","paper","id","paper_W4381786098","section","technologies","entity",{"_15":16,"_29":30,"_31":-5,"_32":-5,"_33":-5,"_34":-5,"_35":36,"_37":38,"_39":-5,"_40":41,"_42":43,"_44":45,"_46":47,"_48":-5,"_49":50,"_51":30,"_52":53,"_54":55,"_56":57,"_58":59,"_60":61,"_62":-5,"_63":64,"_65":-5,"_66":67,"_68":30,"_69":-5},"canonical","https://www.hugmap.com/entity/paper/paper_W4381786098/technologies","siteBaseUrl","https://www.hugmap.com","sectionCount",1,"jsonLd","{\"@context\":\"https://schema.org\",\"@graph\":[{\"name\":\"XSimGCL: Towards Extremely Simple Graph Contrastive Learning for Recommendation\",\"url\":\"https://www.hugmap.com/entity/paper/paper_W4381786098/technologies\",\"@type\":\"ScholarlyArticle\",\"headline\":\"XSimGCL: Towards Extremely Simple Graph Contrastive Learning for Recommendation\",\"datePublished\":\"2023-01-01\",\"isPartOf\":{\"@type\":\"Periodical\",\"name\":\"IEEE Transactions on Knowledge and Data Engineering\"}},{\"@type\":\"CollectionPage\",\"name\":\"XSimGCL: Towards Extremely Simple Graph Contrastive Learning for Recommendation 相关技术\",\"description\":\"XSimGCL: Towards Extremely Simple Graph Contrastive Learning for Recommendation 在 HugMap 中的相关技术，共 1 条，含结构化关联信息与详情链接。\",\"url\":\"https://www.hugmap.com/entity/paper/paper_W4381786098/technologies\",\"inLanguage\":\"zh-CN\"},{\"@type\":\"BreadcrumbList\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"首页\",\"item\":\"https://www.hugmap.com/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"论文\",\"item\":\"https://www.hugmap.com/search?type=paper\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"XSimGCL: Towards Extremely Simple Graph Contrastive Learning for Recommendation\",\"item\":\"https://www.hugmap.com/entity/paper/paper_W4381786098\"},{\"@type\":\"ListItem\",\"position\":4,\"name\":\"相关技术\",\"item\":\"https://www.hugmap.com/entity/paper/paper_W4381786098/technologies\"}]}]}","name","XSimGCL: Towards Extremely Simple Graph Contrastive Learning for Recommendation","nameZh","aliases","description","descriptionZh","status","active","externalIds","{\"openalex\":\"W4381786098\",\"doi\":\"10.1109/tkde.2023.3288135\"}","properties","source","openalex","createdAt","2026-05-22T03:54:46.245049037","updatedAt","2026-05-22T03:54:46.245051277","doi","10.1109/tkde.2023.3288135","arxivId","openalexId","W4381786098","title","abstractText","Contrastive learning (CL) has recently been demonstrated critical in improving recommendation performance. The underlying principle of CL-based recommendation models is to ensure the consistency between representations derived from different graph augmentations of the user-item bipartite graph. This self-supervised approach allows for the extraction of general features from raw data, thereby mitigating the issue of data sparsity. Despite the effectiveness of this paradigm, the factors contributing to its performance gains have yet to be fully understood. This paper provides novel insights into the impact of CL on recommendation. Our findings indicate that CL enables the model to learn more evenly distributed user and item representations, which alleviates the prevalent popularity bias and promoting long-tail items. Our analysis also suggests that the graph augmentations, previously considered essential, are relatively unreliable and of limited significance in CL-based recommendation. Based on these findings, we put forward an e X tremely Sim ple G raph C ontrastive L earning method ( XSimGCL ) for recommendation, which discards the ineffective graph augmentations and instead employs a simple yet effective noise-based embedding augmentation to generate views for CL. A comprehensive experimental study on four large and highly sparse benchmark datasets demonstrates that, though the proposed method is extremely simple, it can smoothly adjust the uniformity of learned representations and outperforms its graph augmentation-based counterparts by a large margin in both recommendation accuracy and training efficiency. The code and used datasets are released at https://github.com/Coder-Yu/SELFRec .","publishedDate","2023-01-01","publishedYear",2023,"venue","IEEE Transactions on Knowledge and Data Engineering","citationCount",265,"authors","keywords",[70,71,72],"pdfUrl","topVenue",false,"displayName","displayDescription","Recommender Systems and Techniques","Advanced Graph Neural Networks","Caching and Content Delivery","mutations",[],"queries",[77],{"_78":79,"_80":81,"_82":83,"_84":85},"dehydratedAt",1781386374540,"state",{"_87":88,"_89":26,"_90":79,"_91":-5,"_92":93,"_94":93,"_95":93,"_96":-5,"_97":-5,"_98":67,"_35":99,"_100":101},"queryKey",[86,19,14,16],"queryHash","[\"portal\",\"entity\",\"paper\",\"paper_W4381786098\"]","portal","data",{"_19":20,"_102":103,"_104":105,"_106":107,"_108":109,"_18":110,"_62":111},"dataUpdateCount","dataUpdatedAt","error","errorUpdateCount",0,"errorUpdatedAt","fetchFailureCount","fetchFailureReason","fetchMeta","isInvalidated","success","fetchStatus","idle","entityType","Paper","taxonomies",[178,179,180],"events",[],"articles",[],[163],[112,113,114,115,116,117],{"_15":157,"_29":158,"_31":-5,"_32":-5,"_33":-5,"_34":-5,"_35":36,"_37":159,"_39":-5,"_40":41,"_42":160,"_44":161,"_123":162,"_51":-5,"_125":-5,"_126":-5,"_127":-5,"_60":-5,"_128":-5,"_129":-5,"_130":-5,"_131":-5,"_132":-5,"_68":158,"_69":-5},{"_15":151,"_29":152,"_31":-5,"_32":-5,"_33":-5,"_34":-5,"_35":36,"_37":153,"_39":-5,"_40":41,"_42":154,"_44":155,"_123":156,"_51":-5,"_125":-5,"_126":-5,"_127":-5,"_60":-5,"_128":-5,"_129":-5,"_130":-5,"_131":-5,"_132":-5,"_68":152,"_69":-5},{"_15":145,"_29":146,"_31":-5,"_32":-5,"_33":-5,"_34":-5,"_35":36,"_37":147,"_39":-5,"_40":41,"_42":148,"_44":149,"_123":150,"_51":-5,"_125":-5,"_126":-5,"_127":-5,"_60":-5,"_128":-5,"_129":-5,"_130":-5,"_131":-5,"_132":-5,"_68":146,"_69":-5},{"_15":139,"_29":140,"_31":-5,"_32":-5,"_33":-5,"_34":-5,"_35":36,"_37":141,"_39":-5,"_40":41,"_42":142,"_44":143,"_123":144,"_51":-5,"_125":-5,"_126":-5,"_127":-5,"_60":-5,"_128":-5,"_129":-5,"_130":-5,"_131":-5,"_132":-5,"_68":140,"_69":-5},{"_15":133,"_29":134,"_31":-5,"_32":-5,"_33":-5,"_34":-5,"_35":36,"_37":135,"_39":-5,"_40":41,"_42":136,"_44":137,"_123":138,"_51":-5,"_125":-5,"_126":-5,"_127":-5,"_60":-5,"_128":-5,"_129":-5,"_130":-5,"_131":-5,"_132":-5,"_68":134,"_69":-5},{"_15":118,"_29":119,"_31":-5,"_32":-5,"_33":-5,"_34":-5,"_35":36,"_37":120,"_39":-5,"_40":41,"_42":121,"_44":122,"_123":124,"_51":-5,"_125":-5,"_126":-5,"_127":-5,"_60":-5,"_128":-5,"_129":-5,"_130":-5,"_131":-5,"_132":-5,"_68":119,"_69":-5},"person_A5084564297","Junliang Yu","{\"openalex\":\"A5084564297\",\"orcid\":\"0000-0003-3401-9829\"}","2026-05-15T17:24:23.188237329","2026-05-22T03:54:46.268941246","orcid","0000-0003-3401-9829","affiliation","researchArea","paperCount","homepage","linkedin","twitter","github","hindex","person_A5006669765","Xin Xia","{\"openalex\":\"A5006669765\",\"orcid\":\"0000-0002-6302-3256\"}","2026-05-15T05:19:50.182212729","2026-05-23T09:51:25.209051195","0000-0002-6302-3256","person_A5100461265","Tong Chen","{\"openalex\":\"A5100461265\",\"orcid\":\"0000-0001-7269-146X\"}","2026-05-15T17:24:25.178213826","2026-05-22T03:54:46.319942308","0000-0001-7269-146X","person_A5101414718","Lizhen Cui","{\"openalex\":\"A5101414718\",\"orcid\":\"0000-0002-8262-8883\"}","2026-05-15T17:24:26.176383574","2026-05-22T20:22:11.18843472","0000-0002-8262-8883","person_A5051219382","Quoc Viet Hung Nguyen","{\"openalex\":\"A5051219382\",\"orcid\":\"0000-0002-9687-1315\"}","2026-05-15T17:24:27.177543198","2026-05-23T08:09:25.229875357","0000-0002-9687-1315","person_A5088492734","Hongzhi Yin","{\"openalex\":\"A5088492734\",\"orcid\":\"0000-0003-1395-261X\"}","2026-05-15T17:24:24.177434519","2026-05-23T08:09:25.193703858","0000-0003-1395-261X",{"_15":164,"_29":71,"_31":-5,"_32":-5,"_33":-5,"_34":-5,"_35":36,"_37":165,"_39":-5,"_40":166,"_42":167,"_44":168,"_169":170,"_171":172,"_173":-5,"_174":175,"_176":-5,"_127":-5,"_177":-5,"_68":71,"_69":-5},"tech_oa_T11273","{\"taxonomy_id\":\"tax_rd_T11273\"}","openalex_topics","2026-05-17T09:00:23.332470439","2026-05-17T09:00:23.332471009","techType","architecture","domain","Physical Sciences","proposedYear","maturityLevel","growing","wikipediaUrl","patentCount",{"_15":205,"_29":71,"_31":-5,"_32":-5,"_33":-5,"_34":-5,"_35":36,"_37":206,"_39":-5,"_40":41,"_42":207,"_44":208,"_185":186,"_187":-5,"_188":189,"_190":209,"_192":210,"_194":93,"_195":196,"_197":93,"_198":-5,"_68":71,"_69":-5},{"_15":199,"_29":70,"_31":-5,"_32":-5,"_33":-5,"_34":-5,"_35":36,"_37":200,"_39":-5,"_40":41,"_42":201,"_44":202,"_185":186,"_187":-5,"_188":189,"_190":203,"_192":204,"_194":93,"_195":196,"_197":93,"_198":-5,"_68":70,"_69":-5},{"_15":181,"_29":72,"_31":-5,"_32":-5,"_33":-5,"_34":-5,"_35":36,"_37":182,"_39":-5,"_40":41,"_42":183,"_44":184,"_185":186,"_187":-5,"_188":189,"_190":191,"_192":193,"_194":93,"_195":196,"_197":93,"_198":-5,"_68":72,"_69":-5},"tax_rd_T11478","{\"openalex_topic\":\"T11478\"}","2026-05-14T00:51:54.188921852","2026-05-23T12:37:22.24808323","taxonomyType","RESEARCH_DOMAIN","code","level",3,"parentId","tax_rd_1705","path","Physical Sciences/Computer Science/Computer Networks and Communications/Caching and Content Delivery","sortOrder","isLeaf",true,"entityCount","children","tax_rd_T10203","{\"openalex_topic\":\"T10203\"}","2026-05-14T00:41:27.503279239","2026-05-23T12:35:19.328904116","tax_rd_1710","Physical Sciences/Computer Science/Information Systems/Recommender Systems and Techniques","tax_rd_T11273","{\"openalex_topic\":\"T11273\"}","2026-05-14T00:38:15.327307853","2026-05-23T12:12:07.237772205","tax_rd_1702","Physical Sciences/Computer Science/Artificial Intelligence/Advanced Graph Neural Networks",{"_73":212,"_75":213},[],[214],{"_78":215,"_80":216,"_82":217,"_84":218},1781386373179,{"_87":220,"_89":26,"_90":215,"_91":-5,"_92":93,"_94":93,"_95":93,"_96":-5,"_97":-5,"_98":67,"_35":99,"_100":101},[86,219],"[\"portal\",\"industries\"]","industries",[221,222,223,224,225,226,227,228,229,230,231,232],{"_15":284,"_29":285,"_31":-5,"_32":286,"_33":287,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":236,"_42":-5,"_44":-5,"_185":237,"_187":-5,"_188":26,"_190":238,"_192":288,"_194":93,"_195":67,"_197":93,"_198":-5,"_68":285,"_69":287},{"_15":280,"_29":281,"_31":-5,"_32":-5,"_33":282,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":236,"_42":-5,"_44":-5,"_185":237,"_187":-5,"_188":26,"_190":238,"_192":283,"_194":93,"_195":67,"_197":93,"_198":-5,"_68":281,"_69":282},{"_15":274,"_29":275,"_31":-5,"_32":276,"_33":277,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":236,"_42":-5,"_44":-5,"_185":237,"_187":-5,"_188":26,"_190":238,"_192":278,"_194":93,"_195":67,"_197":93,"_198":-5,"_68":275,"_69":277},{"_15":270,"_29":271,"_31":-5,"_32":-5,"_33":272,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":236,"_42":-5,"_44":-5,"_185":237,"_187":-5,"_188":26,"_190":238,"_192":273,"_194":93,"_195":67,"_197":93,"_198":-5,"_68":271,"_69":272},{"_15":266,"_29":267,"_31":-5,"_32":-5,"_33":268,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":236,"_42":-5,"_44":-5,"_185":237,"_187":-5,"_188":26,"_190":238,"_192":269,"_194":93,"_195":67,"_197":93,"_198":-5,"_68":267,"_69":268},{"_15":260,"_29":261,"_31":-5,"_32":262,"_33":263,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":236,"_42":-5,"_44":-5,"_185":237,"_187":-5,"_188":26,"_190":238,"_192":264,"_194":93,"_195":67,"_197":93,"_198":-5,"_68":261,"_69":263},{"_15":256,"_29":257,"_31":-5,"_32":-5,"_33":258,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":236,"_42":-5,"_44":-5,"_185":237,"_187":-5,"_188":26,"_190":238,"_192":259,"_194":93,"_195":196,"_197":93,"_198":-5,"_68":257,"_69":258},{"_15":252,"_29":253,"_31":-5,"_32":-5,"_33":254,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":236,"_42":-5,"_44":-5,"_185":237,"_187":-5,"_188":26,"_190":238,"_192":255,"_194":93,"_195":67,"_197":93,"_198":-5,"_68":253,"_69":254},{"_15":248,"_29":249,"_31":-5,"_32":-5,"_33":250,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":236,"_42":-5,"_44":-5,"_185":237,"_187":-5,"_188":26,"_190":238,"_192":251,"_194":93,"_195":67,"_197":93,"_198":-5,"_68":249,"_69":250},{"_15":244,"_29":245,"_31":-5,"_32":-5,"_33":246,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":236,"_42":-5,"_44":-5,"_185":237,"_187":-5,"_188":26,"_190":238,"_192":247,"_194":93,"_195":67,"_197":93,"_198":-5,"_68":245,"_69":246},{"_15":240,"_29":241,"_31":-5,"_32":-5,"_33":242,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":236,"_42":-5,"_44":-5,"_185":237,"_187":-5,"_188":26,"_190":238,"_192":243,"_194":93,"_195":67,"_197":93,"_198":-5,"_68":241,"_69":242},{"_15":233,"_29":234,"_31":-5,"_32":-5,"_33":235,"_34":-5,"_35":36,"_37":-5,"_39":-5,"_40":236,"_42":-5,"_44":-5,"_185":237,"_187":-5,"_188":26,"_190":238,"_192":239,"_194":93,"_195":67,"_197":93,"_198":-5,"_68":234,"_69":235},"tax_ind_smartphone","智能手机","Smartphone","manual","INDUSTRY","tax_ind_root","科技行业/智能手机","tax_ind_robot","机器人","Robotics","科技行业/机器人","tax_ind_wearable","智能穿戴","Smart Wearables","科技行业/智能穿戴","tax_ind_energy","新能源与智能制造","New Energy & Smart Manufacturing","科技行业/新能源与智能制造","tax_ind_biotech","生物科技","Biotech","科技行业/生物科技","tax_ind_quantum","量子计算","Quantum Computing","科技行业/量子计算","tax_ind_blockchain","区块链",[265],"Blockchain","科技行业/区块链","Web3","tax_ind_enterprise","企业软件","Enterprise Software","科技行业/企业软件","tax_ind_security","网络安全","Cybersecurity","科技行业/网络安全","tax_ind_semi","半导体",[279],"Semiconductor","科技行业/半导体","芯片","tax_ind_cloud","云计算","Cloud Computing","科技行业/云计算","tax_ind_ai","人工智能",[289,290],"Artificial Intelligence","科技行业/人工智能","AI","AI 技术","env",{"_293":24},"SITE_BASE_URL"]

XSimGCL: Towards Extremely Simple Graph Contrastive Learning for Recommendation 论文

XSimGCL: Towards Extremely Simple Graph Contrastive Learning for Recommendation · 相关技术

相关技术