{"id":"https://openalex.org/W7139911078","doi":"https://doi.org/10.48550/arxiv.2603.19091","title":"Position: Spectral GNNs Are Neither Spectral Nor Superior for Node Classification","display_name":"Position: Spectral GNNs Are Neither Spectral Nor Superior for Node Classification","publication_year":2026,"publication_date":"2026-03-19","ids":{"openalex":"https://openalex.org/W7139911078","doi":"https://doi.org/10.48550/arxiv.2603.19091"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.19091","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.19091","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","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.2603.19091","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5130218098","display_name":"Qin Jiang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiang, Qin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082176355","display_name":"Chengjia Wang","orcid":"https://orcid.org/0000-0003-2345-7364"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Chengjia","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114658789","display_name":"Michael Lones","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lones, Michael","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130236468","display_name":"Dongdong Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Dongdong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130248694","display_name":"Wei Pang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pang, Wei","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/T11273","display_name":"Advanced Graph Neural Networks","score":0.98580002784729,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11273","display_name":"Advanced Graph Neural Networks","score":0.98580002784729,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T13702","display_name":"Machine Learning in Healthcare","score":0.002099999925121665,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12292","display_name":"Graph Theory and Algorithms","score":0.0008999999845400453,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/laplacian-matrix","display_name":"Laplacian matrix","score":0.5486999750137329},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.507099986076355},{"id":"https://openalex.org/keywords/spectral-method","display_name":"Spectral method","score":0.44690001010894775},{"id":"https://openalex.org/keywords/bandlimiting","display_name":"Bandlimiting","score":0.4465999901294708},{"id":"https://openalex.org/keywords/spectral-graph-theory","display_name":"Spectral graph theory","score":0.41609999537467957},{"id":"https://openalex.org/keywords/fourier-transform","display_name":"Fourier transform","score":0.4138000011444092},{"id":"https://openalex.org/keywords/spectral-properties","display_name":"Spectral properties","score":0.4074000120162964},{"id":"https://openalex.org/keywords/eigenvalues-and-eigenvectors","display_name":"Eigenvalues and eigenvectors","score":0.3806999921798706}],"concepts":[{"id":"https://openalex.org/C115178988","wikidata":"https://www.wikidata.org/wiki/Q772067","display_name":"Laplacian matrix","level":3,"score":0.5486999750137329},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.5123000144958496},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.507099986076355},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.49959999322891235},{"id":"https://openalex.org/C23463724","wikidata":"https://www.wikidata.org/wiki/Q2308831","display_name":"Spectral method","level":2,"score":0.44690001010894775},{"id":"https://openalex.org/C129997835","wikidata":"https://www.wikidata.org/wiki/Q806263","display_name":"Bandlimiting","level":3,"score":0.4465999901294708},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.42480000853538513},{"id":"https://openalex.org/C74003402","wikidata":"https://www.wikidata.org/wiki/Q3180727","display_name":"Spectral graph theory","level":5,"score":0.41609999537467957},{"id":"https://openalex.org/C102519508","wikidata":"https://www.wikidata.org/wiki/Q6520159","display_name":"Fourier transform","level":2,"score":0.4138000011444092},{"id":"https://openalex.org/C2985906921","wikidata":"https://www.wikidata.org/wiki/Q190524","display_name":"Spectral properties","level":2,"score":0.4074000120162964},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.4025999903678894},{"id":"https://openalex.org/C158693339","wikidata":"https://www.wikidata.org/wiki/Q190524","display_name":"Eigenvalues and eigenvectors","level":2,"score":0.3806999921798706},{"id":"https://openalex.org/C62611344","wikidata":"https://www.wikidata.org/wiki/Q1062658","display_name":"Node (physics)","level":2,"score":0.3718000054359436},{"id":"https://openalex.org/C55294529","wikidata":"https://www.wikidata.org/wiki/Q579544","display_name":"Vandermonde matrix","level":3,"score":0.37119999527931213},{"id":"https://openalex.org/C2983668108","wikidata":"https://www.wikidata.org/wiki/Q280453","display_name":"Spectral analysis","level":3,"score":0.35249999165534973},{"id":"https://openalex.org/C140532419","wikidata":"https://www.wikidata.org/wiki/Q249748","display_name":"Spectral radius","level":3,"score":0.335999995470047},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2912999987602234},{"id":"https://openalex.org/C190470478","wikidata":"https://www.wikidata.org/wiki/Q2370229","display_name":"Invariant (physics)","level":2,"score":0.2888999879360199},{"id":"https://openalex.org/C165700671","wikidata":"https://www.wikidata.org/wiki/Q203484","display_name":"Laplace operator","level":2,"score":0.2815999984741211},{"id":"https://openalex.org/C2780069185","wikidata":"https://www.wikidata.org/wiki/Q7977945","display_name":"Equivalence (formal languages)","level":2,"score":0.27559998631477356},{"id":"https://openalex.org/C64339825","wikidata":"https://www.wikidata.org/wiki/Q722659","display_name":"Graph property","level":5,"score":0.2727000117301941},{"id":"https://openalex.org/C207864730","wikidata":"https://www.wikidata.org/wiki/Q179467","display_name":"Fourier series","level":2,"score":0.26910001039505005},{"id":"https://openalex.org/C106937863","wikidata":"https://www.wikidata.org/wiki/Q7236518","display_name":"Power graph analysis","level":3,"score":0.265500009059906},{"id":"https://openalex.org/C203024314","wikidata":"https://www.wikidata.org/wiki/Q1365258","display_name":"Fourier analysis","level":3,"score":0.2644999921321869},{"id":"https://openalex.org/C184720557","wikidata":"https://www.wikidata.org/wiki/Q7825049","display_name":"Topology (electrical circuits)","level":2,"score":0.2535000145435333}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.19091","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.19091","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.19091","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.19091","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","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":{"Spectral":[0,45,173],"Graph":[1,121],"Neural":[2],"Networks":[3],"(Spectral":[4],"GNNs)":[5],"for":[6,64,170],"node":[7,171],"classification":[8],"promise":[9],"frequency-domain":[10],"filtering":[11],"on":[12,16],"graphs,":[13],"yet":[14,106],"rest":[15],"flawed":[17],"foundations.":[18],"Recent":[19],"work":[20],"shows":[21],"that":[22,109,148],"graph":[23,65,179],"Laplacian":[24],"eigenvectors":[25],"do":[26],"not":[27,60,92,141],"in":[28],"general":[29],"have":[30],"the":[31,41,86,158,178],"key":[32],"properties":[33],"of":[34,44,73,97],"a":[35,82],"true":[36],"Fourier":[37,57,62,122],"basis,":[38],"but":[39],"leaves":[40],"empirical":[42],"success":[43],"GNNs":[46,174],"unexplained.":[47],"We":[48,126],"identify":[49],"two":[50,129],"theoretical":[51],"glitches:":[52],"(1)":[53],"commonly":[54,100],"used":[55],"\"graph":[56],"bases\"":[58],"are":[59,187],"classical":[61],"bases":[63],"signals;":[66],"(2)":[67],"(n-1)-degree":[68],"polynomials":[69],"(n":[70],"=":[71],"number":[72],"nodes)":[74],"can":[75],"exactly":[76],"interpolate":[77],"any":[78],"spectral":[79,103,124,132,160],"response":[80],"via":[81],"Vandermonde":[83],"system,":[84],"so":[85],"usual":[87],"\"polynomial":[88],"approximation\"":[89],"narrative":[90],"is":[91,99,140],"theoretically":[93],"justified.":[94],"The":[95],"effectiveness":[96,139],"GCN":[98],"attributed":[101],"to":[102,151,193],"low-pass":[104],"filtering,":[105],"we":[107],"prove":[108],"low-":[110],"and":[111,135],"high-pass":[112],"behaviors":[113],"arise":[114],"solely":[115],"from":[116,145],"message-passing":[117],"dynamics":[118],"rather":[119],"than":[120],"Transform-based":[123],"formulations.":[125],"then":[127],"analyze":[128],"representative":[130],"directed":[131],"models,":[133],"MagNet":[134],"HoloNet.":[136],"Their":[137],"reported":[138],"spectral:":[142],"it":[143],"arises":[144],"implementation":[146],"issues":[147],"reduce":[149],"them":[150],"powerful":[152],"MPNNs.":[153],"When":[154],"implemented":[155],"consistently":[156],"with":[157,200],"claimed":[159],"algorithms,":[161],"performance":[162],"becomes":[163],"weak.":[164],"This":[165],"position":[166],"paper":[167],"argues":[168],"that:":[169],"classification,":[172],"neither":[175],"meaningfully":[176],"capture":[177],"spectrum":[180],"nor":[181],"reliably":[182],"improve":[183],"performance;":[184],"competitive":[185],"results":[186],"better":[188],"explained":[189],"by":[190,197],"their":[191,201],"equivalence":[192],"MPNNs,":[194],"sometimes":[195],"aided":[196],"implementations":[198],"inconsistent":[199],"intended":[202],"design.":[203]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-03-21T00:00:00"}