{"id":"https://openalex.org/W7141099699","doi":"https://doi.org/10.48550/arxiv.2603.25107","title":"Label What Matters: Modality-Balanced and Difficulty-Aware Multimodal Active Learning","display_name":"Label What Matters: Modality-Balanced and Difficulty-Aware Multimodal Active Learning","publication_year":2026,"publication_date":"2026-03-26","ids":{"openalex":"https://openalex.org/W7141099699","doi":"https://doi.org/10.48550/arxiv.2603.25107"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.25107","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.25107","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.2603.25107","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5130760026","display_name":"Yuqiao Zeng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zeng, Yuqiao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130721406","display_name":"Xu Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Xu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062309968","display_name":"Tengfei Liang","orcid":"https://orcid.org/0000-0001-8193-3096"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liang, Tengfei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130814458","display_name":"Yiqing Hao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hao, Yiqing","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130722665","display_name":"Yi Jin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jin, Yi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130776501","display_name":"Hui Yu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Hui","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/T12072","display_name":"Machine Learning and Algorithms","score":0.336899995803833,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.336899995803833,"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":0.22169999778270721,"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":0.06960000097751617,"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/modality","display_name":"Modality (human\u2013computer interaction)","score":0.7752000093460083},{"id":"https://openalex.org/keywords/modalities","display_name":"Modalities","score":0.6747000217437744},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.5978999733924866},{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.47859999537467957},{"id":"https://openalex.org/keywords/multimodal-learning","display_name":"Multimodal learning","score":0.44699999690055847},{"id":"https://openalex.org/keywords/sample","display_name":"Sample (material)","score":0.4462999999523163},{"id":"https://openalex.org/keywords/markov-decision-process","display_name":"Markov decision process","score":0.4415000081062317}],"concepts":[{"id":"https://openalex.org/C2780226545","wikidata":"https://www.wikidata.org/wiki/Q6888030","display_name":"Modality (human\u2013computer interaction)","level":2,"score":0.7752000093460083},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7325999736785889},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.7053999900817871},{"id":"https://openalex.org/C2779903281","wikidata":"https://www.wikidata.org/wiki/Q6888026","display_name":"Modalities","level":2,"score":0.6747000217437744},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5978999733924866},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5770999789237976},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.47859999537467957},{"id":"https://openalex.org/C2780660688","wikidata":"https://www.wikidata.org/wiki/Q25052564","display_name":"Multimodal learning","level":2,"score":0.44699999690055847},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.4462999999523163},{"id":"https://openalex.org/C106189395","wikidata":"https://www.wikidata.org/wiki/Q176789","display_name":"Markov decision process","level":3,"score":0.4415000081062317},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.36340001225471497},{"id":"https://openalex.org/C77967617","wikidata":"https://www.wikidata.org/wiki/Q4677561","display_name":"Active learning (machine learning)","level":2,"score":0.35659998655319214},{"id":"https://openalex.org/C2780910867","wikidata":"https://www.wikidata.org/wiki/Q1952416","display_name":"Multimodality","level":2,"score":0.3476000130176544},{"id":"https://openalex.org/C33954974","wikidata":"https://www.wikidata.org/wiki/Q486494","display_name":"Sensor fusion","level":2,"score":0.31049999594688416},{"id":"https://openalex.org/C23224414","wikidata":"https://www.wikidata.org/wiki/Q176769","display_name":"Hidden Markov model","level":2,"score":0.29170000553131104},{"id":"https://openalex.org/C158525013","wikidata":"https://www.wikidata.org/wiki/Q2593739","display_name":"Fusion","level":2,"score":0.25099998712539673}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.25107","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.25107","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.2603.25107","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.25107","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],"learning":[1,32],"integrates":[2],"complementary":[3],"information":[4],"from":[5],"different":[6],"modalities":[7,81],"such":[8],"as":[9,88,111],"image,":[10],"text,":[11],"and":[12,56,82,124,126,132,154,179,192],"audio":[13],"to":[14,29,69,120,171],"improve":[15],"model":[16],"performance,":[17],"but":[18],"its":[19],"success":[20],"relies":[21],"on":[22,176],"large-scale":[23],"labeled":[24],"data,":[25],"which":[26,65,146,163],"is":[27,52],"costly":[28],"obtain.":[30],"Active":[31],"(AL)":[33],"mitigates":[34],"this":[35,93,138],"challenge":[36],"by":[37],"selectively":[38],"annotating":[39],"informative":[40,173],"samples.":[41,174],"In":[42],"multimodal":[43,74,104],"settings,":[44],"many":[45],"approaches":[46],"implicitly":[47],"assume":[48],"that":[49,182],"modality":[50,121,149,193],"importance":[51],"stable":[53],"across":[54],"rounds":[55],"keep":[57],"selection":[58,110],"rules":[59],"fixed":[60],"at":[61],"the":[62,70,77,83,117,127],"fusion":[63,170],"stage,":[64],"leaves":[66],"them":[67],"insensitive":[68],"dynamic":[71],"nature":[72],"of":[73,80,85],"learning,":[75],"where":[76,116],"relative":[78],"value":[79],"difficulty":[84,166],"instances":[86],"shift":[87],"training":[89],"proceeds.":[90],"To":[91],"address":[92],"issue,":[94],"we":[95],"propose":[96],"RL-MBA,":[97],"a":[98,112],"reinforcement-learning":[99],"framework":[100],"for":[101,158],"modality-balanced,":[102],"difficulty-aware":[103],"active":[105],"learning.":[106],"RL-MBA":[107,183],"models":[108],"sample":[109,165],"Markov":[113],"Decision":[114],"Process,":[115],"policy":[118],"adapts":[119],"contributions,":[122],"uncertainty,":[123],"diversity,":[125],"reward":[128],"encourages":[129],"accuracy":[130,191],"gains":[131],"balance.":[133],"Two":[134],"key":[135],"components":[136],"drive":[137],"adaptability:":[139],"(1)":[140],"Adaptive":[141],"Modality":[142],"Contribution":[143],"Balancing":[144],"(AMCB),":[145],"dynamically":[147],"adjusts":[148],"weights":[150],"via":[151,167],"reinforcement":[152],"feedback,":[153],"(2)":[155],"Evidential":[156],"Fusion":[157],"DifficultyAware":[159],"Policy":[160],"Adjustment":[161],"(EFDA),":[162],"estimates":[164],"uncertainty-based":[168],"evidential":[169],"prioritize":[172],"Experiments":[175],"Food101,":[177],"KineticsSound,":[178],"VGGSound":[180],"demonstrate":[181],"consistently":[184],"outperforms":[185],"strong":[186],"baselines,":[187],"improving":[188],"both":[189],"classification":[190],"fairness":[194],"under":[195],"limited":[196],"labeling":[197],"budgets.":[198]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-03-28T00:00:00"}