{"id":"https://openalex.org/W7161062036","doi":"https://doi.org/10.48550/arxiv.2605.11468","title":"CAMPA: Efficient and Aligned Multimodal Graph Learning via Decoupled Propagation and Aggregation","display_name":"CAMPA: Efficient and Aligned Multimodal Graph Learning via Decoupled Propagation and Aggregation","publication_year":2026,"publication_date":"2026-05-12","ids":{"openalex":"https://openalex.org/W7161062036","doi":"https://doi.org/10.48550/arxiv.2605.11468"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.11468","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.11468","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":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.2605.11468","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5040849589","display_name":"Daohan Su","orcid":"https://orcid.org/0009-0003-1627-9608"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Su, Daohan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136038637","display_name":"Hao Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Hao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136052556","display_name":"Xunkai Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Xunkai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136013036","display_name":"Yinlin Zhu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhu, Yinlin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5094207324","display_name":"XIONG Yongfu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yongfu, Xiong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136075957","display_name":"Yi Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Yi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059652667","display_name":"Hongchao Qin","orcid":"https://orcid.org/0000-0003-4364-0633"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qin, Hongchao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136011897","display_name":"Rong-Hua Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Rong-Hua","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136002181","display_name":"Guoren Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Guoren","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":0.9664999842643738,"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":0.9664999842643738,"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":0.014399999752640724,"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"}},{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.002199999988079071,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.6462000012397766},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.6060000061988831},{"id":"https://openalex.org/keywords/feature-learning","display_name":"Feature learning","score":0.5504999756813049},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.5210999846458435},{"id":"https://openalex.org/keywords/consistency","display_name":"Consistency (knowledge bases)","score":0.39989998936653137},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.3928999900817871},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.3544999957084656},{"id":"https://openalex.org/keywords/limiting","display_name":"Limiting","score":0.35030001401901245}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7670999765396118},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.6462000012397766},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.6060000061988831},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.5504999756813049},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.5210999846458435},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4440999925136566},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.4196000099182129},{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.39989998936653137},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.3928999900817871},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.3544999957084656},{"id":"https://openalex.org/C188198153","wikidata":"https://www.wikidata.org/wiki/Q1613840","display_name":"Limiting","level":2,"score":0.35030001401901245},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.349700003862381},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.34610000252723694},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.3319999873638153},{"id":"https://openalex.org/C174348530","wikidata":"https://www.wikidata.org/wiki/Q188635","display_name":"Bridging (networking)","level":2,"score":0.31130000948905945},{"id":"https://openalex.org/C2780186347","wikidata":"https://www.wikidata.org/wiki/Q11414","display_name":"Subnetwork","level":2,"score":0.2842999994754791},{"id":"https://openalex.org/C2777735758","wikidata":"https://www.wikidata.org/wiki/Q817765","display_name":"Path (computing)","level":2,"score":0.28060001134872437},{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.2687999904155731},{"id":"https://openalex.org/C854659","wikidata":"https://www.wikidata.org/wiki/Q1859284","display_name":"Message passing","level":2,"score":0.26249998807907104},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.2612999975681305},{"id":"https://openalex.org/C43711488","wikidata":"https://www.wikidata.org/wiki/Q7534783","display_name":"Skew","level":2,"score":0.25780001282691956},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.2547999918460846},{"id":"https://openalex.org/C71139939","wikidata":"https://www.wikidata.org/wiki/Q910194","display_name":"Modal","level":2,"score":0.25220000743865967}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.11468","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.11468","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":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.2605.11468","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.11468","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":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":{"Multimodal":[0,111],"Graph":[1],"Neural":[2],"Networks":[3],"(MGNNs)":[4],"have":[5],"shown":[6],"strong":[7,181],"potential":[8],"for":[9,49,116],"learning":[10],"from":[11,26],"multimodal":[12,118],"attributed":[13],"graphs,":[14],"yet":[15],"most":[16],"existing":[17,60],"approaches":[18],"rely":[19],"on":[20,170],"tightly":[21],"coupled":[22,182],"architectures":[23],"that":[24,40,177],"suffer":[25],"prohibitive":[27],"computational":[28],"overhead.":[29],"In":[30],"this":[31,103],"paper,":[32],"we":[33,54,105],"present":[34],"a":[35,56,108,124],"systematic":[36],"empirical":[37],"analysis":[38],"showing":[39],"decoupled":[41,61,117,184,193],"MGNNs":[42],"are":[43],"substantially":[44],"more":[45],"efficient":[46],"and":[47,72,156,160,166,174,183],"scalable":[48],"large-scale":[50],"graph":[51,119],"learning.":[52,100,120],"However,":[53],"identify":[55],"critical":[57],"bottleneck":[58],"in":[59,68],"pipelines,":[62],"namely":[63],"modal":[64],"conflict,":[65],"which":[66,132,152],"arises":[67],"both":[69],"the":[70,188,192],"propagation":[71],"aggregation":[73],"stages.":[74],"Specifically,":[75],"independent":[76],"multi-hop":[77,91],"diffusion":[78],"causes":[79],"cross-modal":[80,129,134],"semantic":[81,142],"divergence":[82],"during":[83,94],"propagation,":[84,131],"while":[85,186],"naive":[86],"fusion":[87],"fails":[88],"to":[89,140,158],"align":[90,161],"feature":[92],"trajectories":[93],"aggregation,":[95,151],"jointly":[96],"limiting":[97],"effective":[98],"representation":[99],"To":[101],"address":[102],"challenge,":[104],"propose":[106],"CAMPA,":[107],"Cross-modal":[109],"Aligned":[110],"Propagation":[112],"&":[113],"Aggregation":[114],"framework":[115],"Concretely,":[121],"CAMPA":[122,178],"introduces":[123],"two-stage":[125],"alignment":[126],"mechanism:":[127],"(1)":[128],"aligned":[130,150],"injects":[133],"similarity":[135],"priors":[136],"into":[137],"message":[138],"passing":[139],"preserve":[141],"consistency":[143],"without":[144],"additional":[145],"parameter":[146],"overhead;":[147],"(2)":[148],"trajectory":[149],"leverages":[153],"trajectory-level":[154],"self-attention":[155],"cross-attention":[157],"capture":[159],"long-range":[162],"dependencies":[163],"across":[164],"modalities":[165],"hops.":[167],"Extensive":[168],"experiments":[169],"diverse":[171],"benchmark":[172],"datasets":[173],"tasks":[175],"demonstrate":[176],"consistently":[179],"outperforms":[180],"baselines":[185],"preserving":[187],"efficiency":[189],"advantages":[190],"of":[191],"paradigm.":[194]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-14T00:00:00"}