[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85632-en":3,"doc-seo-85632-105":30,"detail-sidebar-cat-0-en-105":91},{"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":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},85632,4810365810221,"Aurora","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","RankGraph-2 Lifecycle Co-Design for Billion-Node Graph Learning in Recommendation","RankGraph-2 is a deployed recommendation framework that co-designs graph construction, representation learning, and real-time serving for billion-node similarity-based retrieval. Instead of optimizing stages independently, it links serving cost to a co-learned cluster index, embeds serving-accurate objectives into training, and removes reliance on online graph infrastructure by using pre-computed neighborhoods. Construction supports hour-level refresh for item coverage. RankGraph-2 reduces edge scale via popularity-biased subsampling, pre-computes multi-hop neighborhoods using personalized PageRank, and co-learns residual-quantization indexing to cut serving compute by 83%, yielding higher recall and improved CTR/CVR.","arXiv :2606 . 18379v 3 [ cs .IR] 11 Jul 2026  \nRankGraph-2: Lifecycle Co-Design for Billion-Node Graph Learning in Recommendation  \nRenzhi Wu 1 , Zikun Cui 1 , Junjie Yang 1 , Tai Guo 1 , Hong Li 1 , Xian Chen2 , Li Yu3 , Ke Pan3 , Sri Reddy 1 , Mahesh Srinivasan 1 , Nipun Mathur 1 , Haomin Yu3 , Hong Yan 1  \n1 Meta MRS, 2 Meta Training Data Infra, 3 FB Monetization  \nGraph-based retrieval at billion-node scale requires jointly solving three tightly coupled problems—graph construction, representation learning, and real-time serving—yet existing work addresses each in isolation. We present RankGraph-2, a framework deployed at Meta that co-designs all three lifecycle stages for similarity-based retrieval (U2U2I and U2I2I), where each stage’s requirements shape the others. Serving requires a co-learned cluster index to avoid expensive online KNN—this pushes index co-training into the training objective. Training benefits from the observation that similarity-based retrieval tolerates pre-computed neighborhoods, eliminating online graph infrastructure—this requires construction to produce self-contained data. Construction must also support hour-level refresh for item coverage. Acting on these cascading requirements, RankGraph-2 reduces hundreds of trillions of edges to hundreds of billions via subsampling with popularity bias correction, pre-computes multi-hop neighborhoods via personalized PageRank, and co-learns a residual-quantization cluster index that reduces serving computational cost by 83% . This lifecycle co-design enables a simple architecture to achieve 3.8 × higher recall than a GAT + Deep Graph Infomax model on a bipartite graph and 2.1 × higher than PyTorch-BigGraph on item retrieval. RankGraph-2 delivers up to +0 .96% CTR and +2 .75% CVR, and has powered 20+ retrieval launches across major surfaces.  \nDate: July 14, 2026  \nCorrespondence: Renzhi Wu at [renzhiwu@meta.com](renzhiwu@meta.com)  \nKeywords: Recommendation system, graph learning, graph neural network  \n1 Introduction  \nGraph-based retrieval—using learned embeddings on user-item graphs for candidate generation—is a critical component of modern recommendation systems. Deploying graph neural networks (GNNs) for this purpose at billion-node scale requires solving three tightly coupled problems: constructing the graph, learning representations on it, and serving those representations in real time. These three stages form alifecycle, and a bottleneck in any one stage limits the entire system. Yet existing work overwhelmingly focuses on a single stage in isolation. Academic GNN research optimizes model architectures on small, static graphs (Zhang et al. , 2019 ; Chen et al. , 2023 ; He et al. , 2020) . Industrial systems such as GiGL (Zhao et al. , 2025), LiGNN (Borisyuk et al. , 2024), and GraphScale (Gupta et al. , 2024) advance training infrastructure but assume the input graph is given and do not address serving cost. Neither community has systematically addressed the question: how should graph construction, training, and serving be co-designed when the graph has billions of nodes and  \nhundreds of trillions of potential edges?  \nThis gap matters for retrieval in practice. At Meta, naively constructing the full co-engagement graph is infeasible; naively training on it requires expensive online graph infrastructure; and naively serving the resulting embeddings via online KNN requires thousands of machines. Each stage imposes requirements on the others: graph construction determines what signals training can learn from, the training objective determines what the embeddings capture, and serving cost determines whether the system can be deployed at all. Optimizing any one stage without considering the others leads to suboptimal outcomes.  \nWe present RankGraph-2, a unified framework built on the principle that lifecycle co-design—letting each stage’s requirements shape the others—unlocks performance that optimizing stages independently cannot achieve. RankGraph-2 ","cbCaifhZB7bTG2qE","https://ap.wps.com/l/cbCaifhZB7bTG2qE","pdf",539195,4,1,13,"English","en",105,"# Introduction\n# RankGraph-2 Overview\n## Serving: from Online KNN to Cluster-Based Retrieval\n## Training: Co-Learning for Serving\n## Graph Construction Requirements","[{\"question\":\"What three tightly coupled problems does RankGraph-2 address for billion-node graph retrieval?\",\"answer\":\"It jointly solves graph construction, representation learning, and real-time serving for similarity-based retrieval at billion-node scale.\"},{\"question\":\"Why does RankGraph-2 move away from online KNN during serving?\",\"answer\":\"Online KNN over a constantly changing candidate set is too expensive, so it uses a cluster-based retrieval mechanism backed by an index.\"},{\"question\":\"How does RankGraph-2 reduce serving computation and improve retrieval quality?\",\"answer\":\"It co-trains a residual-quantization cluster index to match KNN accuracy, pre-computes multi-hop neighborhoods via personalized PageRank, and uses popularity-biased subsampling with correction to scale edge processing efficiently.\"}]",1784205125,33,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"rankgraph-2-lifecycle-co-design-for-billion-node-graph-learning-in-recommendation","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":20},"https://docshare.wps.com/document/rankgraph-2-lifecycle-co-design-for-billion-node-graph-learning-in-recommendation/85632/",{"url":52,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-24","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What three tightly coupled problems does RankGraph-2 address for billion-node graph retrieval?","Question",{"text":75,"@type":76},"It jointly solves graph construction, representation learning, and real-time serving for similarity-based retrieval at billion-node scale.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Why does RankGraph-2 move away from online KNN during serving?",{"text":80,"@type":76},"Online KNN over a constantly changing candidate set is too expensive, so it uses a cluster-based retrieval mechanism backed by an index.",{"name":82,"@type":73,"acceptedAnswer":83},"How does RankGraph-2 reduce serving computation and improve retrieval quality?",{"text":84,"@type":76},"It co-trains a residual-quantization cluster index to match KNN accuracy, pre-computes multi-hop neighborhoods via personalized PageRank, and uses popularity-biased subsampling with correction to scale edge processing efficiently.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]