[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85823-en":3,"doc-seo-85823-105":29,"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":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},85823,8796095461564,"Liam","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","SALT-GNN Handling Dense Neighborhoods in Anti-Money Laundering Graphs via Statistics-Aware Attention","Money laundering threatens financial stability and exposes institutions to penalties, motivating automated detection. Laundering schemes often manifest as relational patterns, driving the use of graph neural networks (GNNs) for anti-money laundering (AML). Yet common AML evaluations rely on aggregate metrics that mask operational weaknesses in high-activity recipient contexts. A recipient-degree stratified evaluation shows consistent performance degradation in dense recipient regions across three datasets, traced to multiset non-discriminability, cardinality blindness, and attention fragility under normalization. SALT-GNN combines degree-aware statistical aggregation with attention, improving dense-context F1 while using fewer parameters.","SALT-GNN: Handling Dense Neighborhoods in Anti-Money Laundering Graphs  \nvia Statistics-Aware Attention  \nLidia Losavio, Francesco Sovrano, Dario Fenoglio, Martin Gjoreski, Marc Langheinrich  \nUniversità della Svizzera italiana (USI)  \nLugano, Switzerland  \narXiv :2607 . 10 13 1v 1 [ cs .LG] 11 Jul 2026  \nAbstract  \nMoney laundering threatens financial stability and exposes institutions to penalties, motivating automated detection. Because laundering schemes often emerge through relational patterns, graph neural networks (GNNs) are increasingly used for anti-money laundering (AML) . Yet AML GNNs are typically evaluated with aggregate metrics such as overall F1 score, which hide an operational issue: high-activity recipient accounts concentrate many incoming transactions, making suspicious signals harder to isolate and costlier to investigate. We introduce a recipient-degree stratified evaluation that reports standard AML metrics across recipient-context density. Across three datasets (HI-Small, HI-Medium, and AMLSim-32k-5%), the evaluation reveals consistent degradation in dense recipient contexts, which we trace to three GNN characteristics: the first two are known GNN limitations that AML amplifies, i.e.,(1) multiset non-discriminability and (2) cardinality blindness; and (3) an attention-specific effect, i.e., in dense neighborhoods, normalized attention attenuates weak but pattern-relevant multi-hop signals. Guided by this diagnosis, we propose SALT-GNN, a lightweight statistics-aware architecture that fuses degree-aware statistical aggregation with attention at each message-passing layer. This layer-wise fusion allows distributional and cardinality information to shape the node states used by subsequent attention steps. Ablation results support fusion placement as a key factor in dense-context performance. On HI-Small and HI-Medium it uses up to 77% fewer parameters than task-specific graph-transformer baselines while improving dense-context F1 score by 3–6 points; on AMLSim-32k-5%, it improves highest-degree F1 score by 16–20 points. The gains hold for both Transformer-and GATstyle attention, indicating that the benefit comes from where statistical and attentional evidence is fused rather than from a specific attention operator.  \n1 Introduction  \nMoney laundering conceals illicit asset origins through placement, layering, and integration, involving an estimated 2–5% of global GDP ($800 billion to $2 trillion annually)(United Nations Office on Drugs and Crime 2011; Lannoo and Parlour 2021) . Because compliance failures expose financial institutions to substantial penalties, effective antimoney laundering (AML) systems are operationally essential (Chen et al. 2018). A central difficulty is that laundering rarely appears as a single anomalous transaction (Altman  \nFigure 1: Recipient-context example. A suspicious incoming transaction (red) is easier to isolate at low in-degree (left) than among many normal transfers at high in-degree (right) .  \net al. 2023) . Instead, risk signals often emerge from relational structures such as cycles, fan-in/fan-out, and layered flows (Appendix A) . AML data is therefore naturally graphstructured, with accounts as nodes and transactions as directed edges. Graph Neural Networks (GNNs), which propagate and aggregate information over transaction graphs, have consequently become a prominent modeling family in AML research (Altman et al. 2023; Egressy et al. 2024) . Although engineered-feature pipelines remain competitive (Blanuša et al. 2024), recent AML-tailored GNNs provide strong benchmarks for both transaction-level edge classification and account-level node classification (Suzumura and Kanezashi 2021; Lin et al. 2024) . However, current AML-GNN evaluation remains incomplete. Aggregate metrics such as F1 score and PR-AUC indicate whether a model performs well on average, but they do not show whether it remains reliable in the recipient contexts where many incoming transfers compete to be","cbCaieGtnVS0On7Z","https://ap.wps.com/l/cbCaieGtnVS0On7Z","pdf",6502679,1,19,"English","en",105,"# Abstract\n# Introduction\n## Recipient-degree stratified evaluation\n## Graph-structured AML and GNN limitations\n## Motivation for statistics-aware attention","[{\"question\":\"Why are aggregate metrics like overall F1 insufficient for evaluating AML GNNs?\",\"answer\":\"They summarize performance across all cases and hide whether models remain reliable in recipient contexts where many incoming transfers compete, which is operationally costly to mis-handle.\"},{\"question\":\"What does recipient-degree stratified evaluation reveal about GNN performance?\",\"answer\":\"Across HI-Small, HI-Medium, and AMLSim-32k-5%, performance consistently degrades in dense recipient contexts (high in-degree receivers), even when training and model selection are unchanged.\"},{\"question\":\"How does SALT-GNN improve performance in dense recipient contexts?\",\"answer\":\"It introduces a lightweight statistics-aware architecture that fuses degree-aware statistical aggregation with attention at each message-passing layer, letting distributional and cardinality information shape node states used by subsequent attention steps.\"}]",1784206480,48,{"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":86,"head_meta":88,"extra_data":90,"updated_unix":27},"salt-gnn-handling-dense-neighborhoods-in-anti-money-laundering-graphs-via-statistics-aware-attention","",{"@graph":35,"@context":85},[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/salt-gnn-handling-dense-neighborhoods-in-anti-money-laundering-graphs-via-statistics-aware-attention/85823/",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",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why are aggregate metrics like overall F1 insufficient for evaluating AML GNNs?","Question",{"text":75,"@type":76},"They summarize performance across all cases and hide whether models remain reliable in recipient contexts where many incoming transfers compete, which is operationally costly to mis-handle.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What does recipient-degree stratified evaluation reveal about GNN performance?",{"text":80,"@type":76},"Across HI-Small, HI-Medium, and AMLSim-32k-5%, performance consistently degrades in dense recipient contexts (high in-degree receivers), even when training and model selection are unchanged.",{"name":82,"@type":73,"acceptedAnswer":83},"How does SALT-GNN improve performance in dense recipient contexts?",{"text":84,"@type":76},"It introduces a lightweight statistics-aware architecture that fuses degree-aware statistical aggregation with attention at each message-passing layer, letting distributional and cardinality information shape node states used by subsequent attention steps.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":45,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":21,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":106,"slug":137},"General","general"]