{"id":"https://openalex.org/W7138326150","doi":"https://doi.org/10.1609/aaai.v40i25.39267","title":"Reconcile Gradient Modulation for Harmony Multimodal Learning","display_name":"Reconcile Gradient Modulation for Harmony Multimodal Learning","publication_year":2026,"publication_date":"2026-03-14","ids":{"openalex":"https://openalex.org/W7138326150","doi":"https://doi.org/10.1609/aaai.v40i25.39267"},"language":null,"primary_location":{"id":"doi:10.1609/aaai.v40i25.39267","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i25.39267","pdf_url":"https://ojs.aaai.org/index.php/AAAI/article/download/39267/43228","source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://ojs.aaai.org/index.php/AAAI/article/download/39267/43228","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101560145","display_name":"Xiyuan Gao","orcid":"https://orcid.org/0000-0003-0870-6721"},"institutions":[{"id":"https://openalex.org/I4210131919","display_name":"Xi'an University of Technology","ror":"https://ror.org/038avdt50","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210131919"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Xiyuan Gao","raw_affiliation_strings":["Tianjin University\nXiong'an National Innovation Center\nXiong'an Guochuang Lantian Technology Co., Ltd"],"affiliations":[{"raw_affiliation_string":"Tianjin University\nXiong'an National Innovation Center\nXiong'an Guochuang Lantian Technology Co., Ltd","institution_ids":["https://openalex.org/I4210131919"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128198518","display_name":"Bing Cao","orcid":null},"institutions":[{"id":"https://openalex.org/I4210164952","display_name":"Tianjin haihe hospital","ror":"https://ror.org/05m762q77","country_code":"CN","type":"healthcare","lineage":["https://openalex.org/I4210164952"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bing Cao","raw_affiliation_strings":["Tianjin University\nHaihe Lab of ITAI"],"affiliations":[{"raw_affiliation_string":"Tianjin University\nHaihe Lab of ITAI","institution_ids":["https://openalex.org/I4210164952"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129714010","display_name":"Baoquan Gong","orcid":null},"institutions":[{"id":"https://openalex.org/I162868743","display_name":"Tianjin University","ror":"https://ror.org/012tb2g32","country_code":"CN","type":"education","lineage":["https://openalex.org/I162868743"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Baoquan Gong","raw_affiliation_strings":["Tianjin University"],"affiliations":[{"raw_affiliation_string":"Tianjin University","institution_ids":["https://openalex.org/I162868743"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5129688204","display_name":"Pengfei Zhu","orcid":null},"institutions":[{"id":"https://openalex.org/I4210131919","display_name":"Xi'an University of Technology","ror":"https://ror.org/038avdt50","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210131919"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Pengfei Zhu","raw_affiliation_strings":["Tianjin University\nXiong'an National Innovation Center\nXiong'an Guochuang Lantian Technology Co., Ltd"],"affiliations":[{"raw_affiliation_string":"Tianjin University\nXiong'an National Innovation Center\nXiong'an Guochuang Lantian Technology Co., Ltd","institution_ids":["https://openalex.org/I4210131919"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5101560145"],"corresponding_institution_ids":["https://openalex.org/I4210131919"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.66325037,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"40","issue":"25","first_page":"21225","last_page":"21233"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10812","display_name":"Human Pose and Action Recognition","score":0.18240000307559967,"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"}},"topics":[{"id":"https://openalex.org/T10812","display_name":"Human Pose and Action Recognition","score":0.18240000307559967,"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/T10667","display_name":"Emotion and Mood Recognition","score":0.16210000216960907,"subfield":{"id":"https://openalex.org/subfields/3205","display_name":"Experimental and Cognitive Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.11129999905824661,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/modalities","display_name":"Modalities","score":0.5794000029563904},{"id":"https://openalex.org/keywords/orthogonality","display_name":"Orthogonality","score":0.47130000591278076},{"id":"https://openalex.org/keywords/harmony","display_name":"Harmony (color)","score":0.4675000011920929},{"id":"https://openalex.org/keywords/modality","display_name":"Modality (human\u2013computer interaction)","score":0.4415000081062317},{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.427700012922287},{"id":"https://openalex.org/keywords/energy","display_name":"Energy (signal processing)","score":0.3797999918460846},{"id":"https://openalex.org/keywords/uniqueness","display_name":"Uniqueness","score":0.337799996137619}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6611999869346619},{"id":"https://openalex.org/C2779903281","wikidata":"https://www.wikidata.org/wiki/Q6888026","display_name":"Modalities","level":2,"score":0.5794000029563904},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5516999959945679},{"id":"https://openalex.org/C17137986","wikidata":"https://www.wikidata.org/wiki/Q215067","display_name":"Orthogonality","level":2,"score":0.47130000591278076},{"id":"https://openalex.org/C2776453491","wikidata":"https://www.wikidata.org/wiki/Q5659234","display_name":"Harmony (color)","level":2,"score":0.4675000011920929},{"id":"https://openalex.org/C2780226545","wikidata":"https://www.wikidata.org/wiki/Q6888030","display_name":"Modality (human\u2013computer interaction)","level":2,"score":0.4415000081062317},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.427700012922287},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.3797999918460846},{"id":"https://openalex.org/C2777021972","wikidata":"https://www.wikidata.org/wiki/Q22976830","display_name":"Uniqueness","level":2,"score":0.337799996137619},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.3368000090122223},{"id":"https://openalex.org/C72169020","wikidata":"https://www.wikidata.org/wiki/Q194404","display_name":"Monotonic function","level":2,"score":0.32839998602867126},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.30559998750686646},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2815000116825104},{"id":"https://openalex.org/C167879884","wikidata":"https://www.wikidata.org/wiki/Q727568","display_name":"Balanced flow","level":2,"score":0.2802000045776367},{"id":"https://openalex.org/C2164484","wikidata":"https://www.wikidata.org/wiki/Q5170150","display_name":"Core (optical fiber)","level":2,"score":0.27309998869895935},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.2590000033378601},{"id":"https://openalex.org/C115680565","wikidata":"https://www.wikidata.org/wiki/Q5977448","display_name":"Gradient method","level":2,"score":0.25189998745918274}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1609/aaai.v40i25.39267","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i25.39267","pdf_url":"https://ojs.aaai.org/index.php/AAAI/article/download/39267/43228","source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1609/aaai.v40i25.39267","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i25.39267","pdf_url":"https://ojs.aaai.org/index.php/AAAI/article/download/39267/43228","source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320323993","display_name":"Natural Science Foundation of Tianjin City","ror":null}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7138326150.pdf","grobid_xml":"https://content.openalex.org/works/W7138326150.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Multimodal":[0],"learning":[1],"frequently":[2],"faces":[3],"two":[4],"coupled":[5,66],"challenges:":[6],"modality":[7,17],"imbalance,":[8,58],"where":[9,19],"dominant":[10],"modalities":[11,108],"suppress":[12],"others":[13],"during":[14],"training,":[15],"and":[16,38,56,82,111,123,143,152,164],"conflict,":[18],"opposing":[20],"gradient":[21,44,80],"directions":[22],"hinder":[23],"optimization.":[24],"Existing":[25],"methods":[26],"typically":[27],"address":[28,64],"these":[29],"issues":[30],"in":[31,42,117],"isolation,":[32],"yet":[33],"they":[34],"are":[35,109,170],"intrinsically":[36],"correlated":[37],"most":[39],"fundamentally":[40],"reflected":[41],"the":[43],"space\u2014severe":[45],"imbalance":[46,175],"may":[47,53],"obscure":[48],"conflicts,":[49],"while":[50],"suppressing":[51],"conflict":[52,118,177],"homogenize":[54],"features":[55],"worsen":[57],"affecting":[59],"fusion":[60],"performance.":[61],"To":[62,126],"jointly":[63],"this":[65,128],"challenge,":[67],"we":[68,130],"propose":[69,131],"Reconcile":[70],"Gradient":[71,133],"Modulation":[72],"(RGM),":[73],"a":[74,137,145],"unified":[75],"framework":[76],"that":[77,169],"adaptively":[78],"adjusts":[79],"magnitude":[81],"direction":[83],"for":[84,149],"harmony":[85],"multimodal":[86,174],"learning.":[87,125],"The":[88],"core":[89],"of":[90,140,154],"RGM":[91],"is":[92],"SynOrth":[93],"Grad,":[94],"which":[95],"minimizes":[96],"Dirichlet":[97],"energy":[98],"to":[99,114,172],"perform":[100],"minimal-gradient":[101],"surgery.":[102],"It":[103],"enhances":[104],"cooperation":[105],"synergy":[106],"when":[107],"aligned":[110],"enforces":[112],"orthogonality":[113],"preserve":[115],"uniqueness":[116],"situations,":[119],"thus":[120],"promoting":[121],"stable":[122],"balanced":[124],"guide":[127],"modulation,":[129],"Cumulative":[132],"Energy":[134],"(CGE)":[135],"as":[136],"convergence-guaranteed":[138],"measure":[139],"modality-wise":[141],"progress,":[142],"construct":[144],"Balance-nonConflict":[146],"Plane":[147],"(BCP)":[148],"real-time":[150],"diagnosis":[151],"control":[153],"training":[155],"dynamics.":[156],"Experiments":[157],"on":[158],"diverse":[159],"benchmarks":[160],"validate":[161],"our":[162],"effectiveness":[163],"generalizability,":[165],"consistently":[166],"outperforming":[167],"counterparts":[168],"designed":[171],"handle":[173],"or":[176],"independently.":[178]},"counts_by_year":[],"updated_date":"2026-04-21T08:09:41.155169","created_date":"2026-03-18T00:00:00"}