[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83617-en":3,"doc-seo-83617-105":30,"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":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},83617,16904993612988,"Olivia Brown","https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd",8,"Research & Report","SABER A Semantic-Aligned Brain Network Analysis Framework via Multi-scale Hypergraphs","Effective brain disease diagnosis depends on combining brain connectivity patterns with high-level semantic knowledge, yet existing approaches treat LLM semantics only as auxiliary features or supervision, which limits stable decision-making and robustness. SABER actively integrates LLM-derived semantics into prediction. ROI-level semantics enrich node representations through global self-attention, multi-scale hypergraphs model functional subnetworks and multi-ROI interactions, and decision-level alignment selectively injects patient-specific textual embeddings to guide classification directly. Experiments on ABIDE and ADHD-200 show state-of-the-art performance, improved stability, and better interpretability, especially in small-sample settings.","SABER: A Semantic-Aligned Brain Network Analysis Framework via Multi-scale Hypergraphs  \nYidan Xu 1 , Xiangmin Han2 , Rundong Xue3 , Huihui Ye 1 ,†  \n1 Hangzhou Dianzi University, China  \n2 Tsinghua University, China  \n3 Xi’an Jiaotong University, China  \n[Emails: YidanXu2024@163.com](Emails: YidanXu2024@163.com), [simon.xmhan@gmail.com](simon.xmhan@gmail.com), [xuerundong@stu.xjtu.edu.cn](xuerundong@stu.xjtu.edu.cn), [yehuihui@hdu.edu.cn](yehuihui@hdu.edu.cn)  \narXiv :2607 .0 190 1v 1 [ cs .LG] 2 Jul 2026  \nAbstract—Effective brain disease diagnosis requires the synergy of brain connectivity patterns and high-level semantic knowledge. Existing methods, however, largely treat semantics from large language models (LLMs) as auxiliary features or supervision, limiting their direct role in decision-making and constraining classification stability and robustness. To overcome this, we propose a semantic-aligned brain network framework that actively integrates LLM-derived semantics into the prediction process. Specifically, ROI-level semantics are first incorporated via global self-attention to enrich node representations and provide whole-brain context. Multi-scale hypergraphs are then constructed to explicitly model functional subnetworks and multiROI interactions, addressing the locality limitations of traditional GNNs and capturing high-order dependencies. Finally, a decision-level semantic alignment mechanism selectively injects patient-specific textual embeddings into graph representations, enabling semantics to directly guide predictions without perturbing the underlying network structure. Experiments on public brain network datasets ABIDE and ADHD-200 demonstrate state-of-the-art performance, enhanced stability, and improved interpretability, particularly in small-sample settings.  \nIndex Terms—Hypergraph Neural Networks, Brain Network Analysis, Semantic Alignment.  \nI. INTRODUCTION  \nFunctional magnetic resonance imaging (fMRI) is a pivotal non-invasive tool for brain network analysis [1]–[3] . By modeling functional connectivity as brain networks, it enables systematic investigation of inter-regional interactions, offering critical diagnostic insights for Autism Spectrum Disorder (ASD) and Attention-Deficit/Hyperactivity Disorder (ADHD) . However, complex high-order dependencies, intersubject variability, and underutilized clinical semantics hinder the extraction of discriminative features.  \nSince brain networks can be naturally modeled as graph structures, GNNs have emerged as a paradigm for brain modeling. Representative methods like BrainGNN [4] and DHGFormer [5] have advanced the analysis of functional connectivity. To further capture complex interactions, hypergraphbased extensions such as I2HGC [6] have been introduced to model high-order correlations among brain regions. Most graph frameworks overlook the alignment of structural inter  \nThis work is supported in part by the Zhejiang Provincial “Jianbing Lingyan+X” Science and Technology Program (2025C01127) . Corresponding author: Huihui Ye ([yehuihui@hdu.edu.cn](yehuihui@hdu.edu.cn)).  \nFig. 1. Overview of the LLM-driven pipeline in our proposed Saber framework for generating ROI-level, patient-level, and disease-level semantics, serving as structured semantic priors for subsequent brain network analysis.  \nactions with clinical semantics, failing to leverage semantic intervention for discriminative learning.  \nIn practice, clinicians interpret neuroimaging evidence in conjunction with prior knowledge, such as functional roles of brain regions and disease-related semantics. The emergence of Large Language Models (LLMs) offers a new avenue to encode such high-level knowledge and incorporate it into brain network analysis [7] . Consequently, synergizing highorder brain network modeling with LLM-derived semantics represents an essential research direction.  \nTwo main paradigms currently dominate the integration offMRI with LLMs. The first involves directly mapping fMR","cbCain0DfqxDpPg1","https://ap.wps.com/l/cbCain0DfqxDpPg1","pdf",1347436,3,1,6,"English","en",105,"# Introduction\n## Challenges in integrating clinical semantics\n## Existing LLM–fMRI integration paradigms\n## Proposed solution and decision-level semantic alignment","[{\"question\":\"What role do multi-scale hypergraphs play in SABER?\",\"answer\":\"Multi-scale hypergraphs explicitly model functional subnetworks and multi-ROI interactions, overcoming locality limits of traditional GNNs and capturing high-order dependencies beyond pairwise edges.\"}]",1784189300,15,{"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":78,"head_meta":80,"extra_data":82,"updated_unix":28},"saber-a-semantic-aligned-brain-network-analysis-framework-via-multi-scale-hypergraphs","",{"@graph":36,"@context":77},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,50],{"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":20},"https://docshare.wps.com/document/research-report/",{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/saber-a-semantic-aligned-brain-network-analysis-framework-via-multi-scale-hypergraphs/83617/",4,{"url":51,"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","2026-07-27","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},"What role do multi-scale hypergraphs play in SABER?","Question",{"text":75,"@type":76},"Multi-scale hypergraphs explicitly model functional subnetworks and multi-ROI interactions, overcoming locality limits of traditional GNNs and capturing high-order dependencies beyond pairwise edges.","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":25},{"code":4,"msg":5,"data":84},[85,89,93,97,102,106,111,114,119,122,126],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":86,"show_sort_weight":87,"slug":88},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":90,"show_sort_weight":91,"slug":92},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Exam",70,"exam",{"id":98,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},5,"Comic",60,"comic",{"id":22,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":112,"slug":113},30,"research-report",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},9,"Religion & Spirituality",20,"religion-spirituality",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":120,"show_sort_weight":117,"slug":121},"World Cup","world-cup",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":123,"slug":125},10,"Lifestyle","lifestyle",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":98,"slug":129},19,"General","general"]