{"id":"https://openalex.org/W7162428736","doi":"https://doi.org/10.48550/arxiv.2605.24684","title":"Beyond the Aggregation Dilemma: Prior-Retaining Decoupled Learning for Multimodal Graphs","display_name":"Beyond the Aggregation Dilemma: Prior-Retaining Decoupled Learning for Multimodal Graphs","publication_year":2026,"publication_date":"2026-05-23","ids":{"openalex":"https://openalex.org/W7162428736","doi":"https://doi.org/10.48550/arxiv.2605.24684"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.24684","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.24684","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":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.24684","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137075922","display_name":"Hao Yan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yan, Hao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137004174","display_name":"Xuanru Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Xuanru","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136995688","display_name":"Jun Yin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yin, Jun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137026468","display_name":"Shirui Pan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pan, Shirui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137082578","display_name":"Senzhang Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Senzhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5137051843","display_name":"Chengqi Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Chengqi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":6,"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":0.9546999931335449,"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":0.9546999931335449,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.014299999922513962,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.0034000000450760126,"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/discriminative-model","display_name":"Discriminative model","score":0.6074000000953674},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.5354999899864197},{"id":"https://openalex.org/keywords/topology","display_name":"Topology (electrical circuits)","score":0.47510001063346863},{"id":"https://openalex.org/keywords/inversion","display_name":"Inversion (geology)","score":0.36809998750686646},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.3479999899864197},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.3441999852657318},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.34290000796318054}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7088000178337097},{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.6074000000953674},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.5354999899864197},{"id":"https://openalex.org/C184720557","wikidata":"https://www.wikidata.org/wiki/Q7825049","display_name":"Topology (electrical circuits)","level":2,"score":0.47510001063346863},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4359000027179718},{"id":"https://openalex.org/C1893757","wikidata":"https://www.wikidata.org/wiki/Q3653001","display_name":"Inversion (geology)","level":3,"score":0.36809998750686646},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.3479999899864197},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.3441999852657318},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.34290000796318054},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.34139999747276306},{"id":"https://openalex.org/C189950617","wikidata":"https://www.wikidata.org/wiki/Q937228","display_name":"Property (philosophy)","level":2,"score":0.3305000066757202},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.32260000705718994},{"id":"https://openalex.org/C62611344","wikidata":"https://www.wikidata.org/wiki/Q1062658","display_name":"Node (physics)","level":2,"score":0.3190000057220459},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.3084000051021576},{"id":"https://openalex.org/C2779903281","wikidata":"https://www.wikidata.org/wiki/Q6888026","display_name":"Modalities","level":2,"score":0.30649998784065247},{"id":"https://openalex.org/C199845137","wikidata":"https://www.wikidata.org/wiki/Q145490","display_name":"Network topology","level":2,"score":0.2971999943256378},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.28049999475479126},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.27970001101493835},{"id":"https://openalex.org/C46686674","wikidata":"https://www.wikidata.org/wiki/Q466303","display_name":"Boosting (machine learning)","level":2,"score":0.262800008058548}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.24684","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.24684","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":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.24684","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.24684","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":"article"},"sustainable_development_goals":[{"score":0.7310999035835266,"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Multimodal":[0,188],"Attributed":[1],"Graph":[2,189],"Learning":[3],"(MAGL)":[4],"integrates":[5],"intrinsic":[6,89],"node":[7],"attributes":[8],"with":[9,91,158],"structural":[10,151],"topology":[11],"via":[12,153],"graph":[13],"aggregation.":[14],"However,":[15],"as":[16],"pretrained":[17],"encoders":[18],"evolve":[19],"into":[20],"Large":[21],"Foundation":[22],"Models":[23],"(LFMs),":[24],"the":[25,95],"landscape":[26],"of":[27],"MAGL":[28,51],"fundamentally":[29],"shifts:":[30],"under":[31],"high-confidence":[32],"LFM":[33],"priors,":[34],"mandatory":[35,85],"aggregation":[36,72,86,111],"introduces":[37],"topological":[38,92,110],"noise":[39,96],"that":[40,65,168],"overwhelms":[41],"discriminative":[42],"signals,":[43],"triggering":[44],"a":[45,70,116,138,154],"counter-intuitive":[46],"performance":[47,172],"inversion":[48,67],"where":[49],"sophisticated":[50],"architectures":[52],"underperform":[53],"simple":[54],"topology-agnostic":[55,147],"MLPs.":[56],"Through":[57],"systematic":[58],"empirical":[59],"and":[60,103,180],"theoretical":[61],"analysis,":[62],"we":[63,132],"identify":[64],"this":[66,130],"stems":[68],"from":[69],"fundamental":[71],"dilemma":[73],"characterized":[74],"by":[75],"two":[76],"concurrent":[77],"pathologies:":[78],"(1)":[79],"Representational":[80],"Pathology":[81,106],"(SNR":[82],"Degradation)":[83],"-":[84,109],"dilutes":[87],"robust":[88],"features":[90,145],"noise,":[93],"causing":[94],"penalty":[97],"to":[98,123,182],"outweigh":[99],"its":[100],"collaborative":[101],"benefit;":[102],"(2)":[104],"Optimization":[105],"(Gradient":[107],"Starvation)":[108],"attenuates":[112],"gradient":[113,163],"flow,":[114],"while":[115,149,173],"shared":[117,156],"task":[118],"loss":[119],"causes":[120],"dominant":[121],"modalities":[122],"prematurely":[124],"suppress":[125],"weaker":[126],"ones.":[127],"To":[128],"resolve":[129],"dilemma,":[131],"propose":[133],"SUPRA":[134,142,169],"(Shared-Unique":[135],"Prior-Retaining":[136],"Architecture),":[137],"decoupled":[139],"dual-pathway":[140],"paradigm.":[141],"processes":[143],"modality-specific":[144],"through":[146],"MLPs":[148],"capturing":[150],"synergy":[152],"lightweight":[155],"GNN,":[157],"auxiliary":[159],"deep":[160],"supervision":[161],"counteracting":[162],"starvation.":[164],"Extensive":[165],"evaluations":[166],"demonstrate":[167],"achieves":[170],"state-of-the-art":[171],"requiring":[174],"3.5x":[175],"lower":[176],"peak":[177],"GPU":[178],"memory":[179],"up":[181],"4.4x":[183],"faster":[184],"training":[185],"time":[186],"than":[187],"Transformers.":[190]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-27T00:00:00"}