{"id":"https://openalex.org/W7169070910","doi":"https://doi.org/10.48550/arxiv.2607.13241","title":"EMAGN: Efficient Multi-Attention Graph Network via Learned Clustering for Scalable Traffic Forecasting","display_name":"EMAGN: Efficient Multi-Attention Graph Network via Learned Clustering for Scalable Traffic Forecasting","publication_year":2026,"publication_date":"2026-07-14","ids":{"openalex":"https://openalex.org/W7169070910","doi":"https://doi.org/10.48550/arxiv.2607.13241"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2607.13241","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.13241","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2607.13241","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5049072744","display_name":"Mei\u2010Yan Xu","orcid":"https://orcid.org/0000-0002-4234-4296"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Mingxing","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077430258","display_name":"Rakesh Chowdary Machineni","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Machineni, Rakesh Chowdary","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5141026352","display_name":"Ke Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Ke","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140977444","display_name":"Xi Cheng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cheng, Xi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071078743","display_name":"Chengqi Lu","orcid":"https://orcid.org/0000-0002-1930-4675"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lu, Chengqi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5141005748","display_name":"Xin Hu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hu, Xin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140987400","display_name":"Lyuhao Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Lyuhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5141003776","display_name":"Xiangyu Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Xiangyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5035231067","display_name":"Junwei You","orcid":"https://orcid.org/0009-0002-6447-8276"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"You, Junwei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5119968696","display_name":"Oliver Gao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gao, Oliver","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11344","display_name":"Traffic Prediction and Management Techniques","score":0.9692000150680542,"subfield":{"id":"https://openalex.org/subfields/2215","display_name":"Building and Construction"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11344","display_name":"Traffic Prediction and Management Techniques","score":0.9692000150680542,"subfield":{"id":"https://openalex.org/subfields/2215","display_name":"Building and Construction"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10524","display_name":"Traffic control and management","score":0.002400000113993883,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T13918","display_name":"Advanced Data and IoT Technologies","score":0.002199999988079071,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.656499981880188},{"id":"https://openalex.org/keywords/cluster-analysis","display_name":"Cluster analysis","score":0.571399986743927},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5310999751091003},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.4839000105857849},{"id":"https://openalex.org/keywords/flexibility","display_name":"Flexibility (engineering)","score":0.4578999876976013},{"id":"https://openalex.org/keywords/quadratic-equation","display_name":"Quadratic equation","score":0.4043000042438507},{"id":"https://openalex.org/keywords/categorical-variable","display_name":"Categorical variable","score":0.391400009393692},{"id":"https://openalex.org/keywords/dependency","display_name":"Dependency (UML)","score":0.3837999999523163}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7993000149726868},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.656499981880188},{"id":"https://openalex.org/C73555534","wikidata":"https://www.wikidata.org/wiki/Q622825","display_name":"Cluster analysis","level":2,"score":0.571399986743927},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5310999751091003},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.4839000105857849},{"id":"https://openalex.org/C2780598303","wikidata":"https://www.wikidata.org/wiki/Q65921492","display_name":"Flexibility (engineering)","level":2,"score":0.4578999876976013},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.45419999957084656},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.41609999537467957},{"id":"https://openalex.org/C129844170","wikidata":"https://www.wikidata.org/wiki/Q41299","display_name":"Quadratic equation","level":2,"score":0.4043000042438507},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3959999978542328},{"id":"https://openalex.org/C5274069","wikidata":"https://www.wikidata.org/wiki/Q2285707","display_name":"Categorical variable","level":2,"score":0.391400009393692},{"id":"https://openalex.org/C19768560","wikidata":"https://www.wikidata.org/wiki/Q320727","display_name":"Dependency (UML)","level":2,"score":0.3837999999523163},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.35339999198913574},{"id":"https://openalex.org/C311688","wikidata":"https://www.wikidata.org/wiki/Q2393193","display_name":"Time complexity","level":2,"score":0.31459999084472656},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.31119999289512634},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3052000105381012},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.29679998755455017},{"id":"https://openalex.org/C158622935","wikidata":"https://www.wikidata.org/wiki/Q660848","display_name":"Nonlinear system","level":2,"score":0.2865000069141388},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.2816999852657318},{"id":"https://openalex.org/C61326573","wikidata":"https://www.wikidata.org/wiki/Q1496376","display_name":"Gaussian process","level":3,"score":0.27149999141693115},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.26339998841285706},{"id":"https://openalex.org/C2776036281","wikidata":"https://www.wikidata.org/wiki/Q48769818","display_name":"Constraint (computer-aided design)","level":2,"score":0.2529999911785126}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2607.13241","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.13241","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2607.13241","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.13241","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Traffic":[0],"forecasting":[1],"is":[2],"highly":[3],"challenging":[4],"due":[5,34],"to":[6,20,35,89,154,180],"complex":[7],"and":[8,11,23,38,73,78,105,128,168,173],"nonlinear":[9],"spatial":[10,55],"temporal":[12],"dependencies.":[13],"Self-attention":[14],"mechanisms":[15],"have":[16],"been":[17],"widely":[18],"adopted":[19],"model":[21,162],"dynamic":[22,98],"long-range":[24],"dependencies,":[25],"achieving":[26],"state-of-the-art":[27],"performance,":[28],"but":[29],"suffer":[30],"from":[31,86],"limited":[32],"scalability":[33],"quadratic":[36],"computational":[37],"memory":[39,130,143],"complexity.":[40],"To":[41],"address":[42],"this,":[43],"we":[44],"propose":[45],"an":[46],"Efficient":[47],"Multi-Attention":[48],"Graph":[49],"Network":[50],"(EMAGN)":[51],"that":[52,108],"linearises":[53],"the":[54,61,93,176],"attention":[56,96,136],"mechanism":[57],"itself,":[58],"inspired":[59],"by":[60,122,126,131],"theory":[62],"of":[63,95,115,142,160],"fast":[64],"high-dimensional":[65],"Gaussian":[66],"filtering.":[67],"Two":[68],"learned":[69],"clustering":[70],"matrices":[71],"C_k":[72],"C_v":[74],"adaptively":[75],"group":[76],"key":[77],"value":[79],"vectors":[80],"into":[81],"M":[82],"super-clusters,":[83],"reducing":[84,119],"complexity":[85],"O(N^2":[87],"d)":[88],"O(NMd)":[90],"without":[91],"sacrificing":[92],"flexibility":[94],"for":[97],"dependency":[99],"modelling.":[100],"Experimental":[101],"results":[102],"on":[103,144],"PEMS-BAY":[104],"METR-LA":[106],"show":[107],"EMAGN":[109,152,164],"achieves":[110],"accuracy":[111,172],"within":[112,175],"2.7-3.2%":[113],"MAE":[114],"full-attention":[116,138],"GMAN":[117,139],"while":[118,151],"training":[120],"time":[121,125],"32%,":[123],"inference":[124],"38%,":[127],"GPU":[129,149],"58%.":[132],"Critically,":[133],"at":[134],"K=16":[135],"heads,":[137],"runs":[140],"out":[141],"a":[145,157],"standard":[146],"11":[147],"GB":[148],"entirely":[150],"continues":[153],"operate,":[155],"demonstrating":[156],"categorical":[158],"expansion":[159],"feasible":[161],"configurations.":[163],"also":[165],"surpasses":[166],"Linformer":[167],"Performer":[169],"in":[170],"both":[171],"efficiency":[174],"same":[177],"backbone,":[178],"owing":[179],"its":[181],"traffic-network-aware":[182],"adaptive":[183],"clustering.":[184]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-07-17T00:00:00"}