{"id":"https://openalex.org/W7164179726","doi":"https://doi.org/10.1109/lsp.2026.3702515","title":"DeRA-MOS: Optimizing Text-to-Music Evaluation via Decoupled Listwise Ranking and Modality Alignment","display_name":"DeRA-MOS: Optimizing Text-to-Music Evaluation via Decoupled Listwise Ranking and Modality Alignment","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7164179726","doi":"https://doi.org/10.1109/lsp.2026.3702515"},"language":null,"primary_location":{"id":"doi:10.1109/lsp.2026.3702515","is_oa":false,"landing_page_url":"https://doi.org/10.1109/lsp.2026.3702515","pdf_url":null,"source":{"id":"https://openalex.org/S120629676","display_name":"IEEE Signal Processing Letters","issn_l":"1070-9908","issn":["1070-9908","1558-2361"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Signal Processing Letters","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5138393010","display_name":"Chien-Chun Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210087940","display_name":"Technology Holding (United States)","ror":"https://ror.org/005mr0595","country_code":"US","type":"company","lineage":["https://openalex.org/I4210087940"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chien-Chun Wang","raw_affiliation_strings":["E.SUN Financial Holding Company, Ltd., Taipei, Taiwan"],"raw_orcid":"https://orcid.org/0009-0000-1392-0058","affiliations":[{"raw_affiliation_string":"E.SUN Financial Holding Company, Ltd., Taipei, Taiwan","institution_ids":["https://openalex.org/I4210087940"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138328388","display_name":"Hung-Shin Lee","orcid":null},"institutions":[{"id":"https://openalex.org/I38007976","display_name":"De Lin Institute of Technology","ror":"https://ror.org/00y3by878","country_code":"TW","type":"education","lineage":["https://openalex.org/I38007976"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Hung-Shin Lee","raw_affiliation_strings":["United Link Company, Ltd., Taipei, Taiwan"],"raw_orcid":"https://orcid.org/0000-0001-7044-9434","affiliations":[{"raw_affiliation_string":"United Link Company, Ltd., Taipei, Taiwan","institution_ids":["https://openalex.org/I38007976"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138387233","display_name":"Hsin-Min Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210098366","display_name":"Institute of Information Science, Academia Sinica","ror":"https://ror.org/00z83z196","country_code":"TW","type":"facility","lineage":["https://openalex.org/I4210098366","https://openalex.org/I84653119"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Hsin-Min Wang","raw_affiliation_strings":["Institute of Information Science, Academia Sinica, Taipei, Taiwan"],"raw_orcid":"https://orcid.org/0000-0003-3599-5071","affiliations":[{"raw_affiliation_string":"Institute of Information Science, Academia Sinica, Taipei, Taiwan","institution_ids":["https://openalex.org/I4210098366"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5138335590","display_name":"Berlin Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I134161618","display_name":"National Taiwan Normal University","ror":"https://ror.org/059dkdx38","country_code":"TW","type":"education","lineage":["https://openalex.org/I134161618"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Berlin Chen","raw_affiliation_strings":["Department of Computer Science and Information Engineering, National Taiwan Normal University, Taipei, Taiwan"],"raw_orcid":"https://orcid.org/0000-0003-0693-8932","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Information Engineering, National Taiwan Normal University, Taipei, Taiwan","institution_ids":["https://openalex.org/I134161618"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.74263019,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"33","issue":null,"first_page":"2740","last_page":"2744"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11309","display_name":"Music and Audio Processing","score":0.9442999958992004,"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"}},"topics":[{"id":"https://openalex.org/T11309","display_name":"Music and Audio Processing","score":0.9442999958992004,"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"}},{"id":"https://openalex.org/T10203","display_name":"Recommender Systems and Techniques","score":0.0038999998942017555,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10664","display_name":"Sentiment Analysis and Opinion Mining","score":0.0038999998942017555,"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/modality","display_name":"Modality (human\u2013computer interaction)","score":0.574999988079071},{"id":"https://openalex.org/keywords/ranking","display_name":"Ranking (information retrieval)","score":0.4611999988555908},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.37290000915527344},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.289900004863739},{"id":"https://openalex.org/keywords/algorithm-design","display_name":"Algorithm design","score":0.2759999930858612}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7329999804496765},{"id":"https://openalex.org/C2780226545","wikidata":"https://www.wikidata.org/wiki/Q6888030","display_name":"Modality (human\u2013computer interaction)","level":2,"score":0.574999988079071},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5479999780654907},{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.4611999988555908},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3779999911785126},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.37290000915527344},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.33500000834465027},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.328900009393692},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3154999911785126},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.289900004863739},{"id":"https://openalex.org/C106516650","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm design","level":2,"score":0.2759999930858612},{"id":"https://openalex.org/C48677424","wikidata":"https://www.wikidata.org/wiki/Q6888088","display_name":"Mode (computer interface)","level":2,"score":0.2741999924182892},{"id":"https://openalex.org/C104267543","wikidata":"https://www.wikidata.org/wiki/Q208163","display_name":"Signal processing","level":3,"score":0.27059999108314514},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.26159998774528503}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/lsp.2026.3702515","is_oa":false,"landing_page_url":"https://doi.org/10.1109/lsp.2026.3702515","pdf_url":null,"source":{"id":"https://openalex.org/S120629676","display_name":"IEEE Signal Processing Letters","issn_l":"1070-9908","issn":["1070-9908","1558-2361"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Signal Processing Letters","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":28,"referenced_works":["https://openalex.org/W9014458","https://openalex.org/W1881549354","https://openalex.org/W2091158010","https://openalex.org/W2108862644","https://openalex.org/W2128877075","https://openalex.org/W2330485005","https://openalex.org/W2949208911","https://openalex.org/W2972394484","https://openalex.org/W2972478942","https://openalex.org/W3097760447","https://openalex.org/W3161558238","https://openalex.org/W3176445421","https://openalex.org/W3196475561","https://openalex.org/W3202278141","https://openalex.org/W3207932315","https://openalex.org/W4225956675","https://openalex.org/W4239181501","https://openalex.org/W4372260310","https://openalex.org/W4385152092","https://openalex.org/W4392902957","https://openalex.org/W4404577625","https://openalex.org/W4408353668","https://openalex.org/W4413472064","https://openalex.org/W4416799148","https://openalex.org/W7133229781","https://openalex.org/W7148359692","https://openalex.org/W7148392549","https://openalex.org/W7148400581"],"related_works":[],"abstract_inverted_index":{"Evaluating":[0],"text-to-music":[1],"(TTM)":[2],"systems":[3],"remains":[4],"expensive":[5],"because":[6],"music":[7],"impression":[8],"(MI)":[9],"and":[10,42,80,109,123,140],"text":[11],"alignment":[12,99],"(TA)":[13],"scores":[14,20,104],"rely":[15],"on":[16,86,127],"human":[17,103],"mean":[18],"opinion":[19],"(MOS).":[21],"Most":[22],"automatic":[23],"MOS":[24],"estimators":[25],"are":[26],"trained":[27],"with":[28,83],"point-wise":[29,120],"regression":[30],"or":[31],"distributional":[32],"classification.":[33],"These":[34],"objectives":[35],"do":[36],"not":[37],"directly":[38],"optimize":[39],"rank-based":[40],"metrics":[41],"provide":[43],"weak":[44],"geometric":[45],"constraints":[46],"for":[47,61,148],"cross-modal":[48],"coherence.":[49],"To":[50],"address":[51],"these":[52],"gaps,":[53],"we":[54,66,94],"propose":[55],"DeRA-MOS,":[56],"a":[57,68,96,145],"decoupled":[58,132],"optimization":[59],"framework":[60,133],"TTM":[62,150],"evaluation.":[63,151],"For":[64,92],"MI,":[65],"introduce":[67,95],"batch-aware":[69],"listwise":[70],"ranking":[71,142],"loss":[72,100],"that":[73,101,130],"models":[74],"relative":[75],"order":[76],"within":[77],"each":[78],"mini-batch":[79],"better":[81],"aligns":[82],"evaluation":[84],"based":[85],"Spearman's":[87],"rank":[88],"correlation":[89],"coefficient":[90],"(SRCC).":[91],"TA,":[93],"score-anchored":[97],"modality":[98,124],"maps":[102],"to":[105],"target":[106],"audio-text":[107],"similarity":[108],"regularizes":[110],"the":[111,119],"latent":[112],"space":[113],"before":[114],"fusion.":[115],"By":[116],"effectively":[117],"mitigating":[118],"training":[121],"mismatch":[122],"drift,":[125],"experiments":[126],"MusicEval":[128],"demonstrate":[129],"our":[131],"yields":[134],"substantial":[135],"improvements":[136],"in":[137],"both":[138],"MI":[139],"TA":[141],"metrics,":[143],"establishing":[144],"robust":[146],"paradigm":[147],"large-scale":[149]},"counts_by_year":[],"updated_date":"2026-07-18T05:51:51.687321","created_date":"2026-06-11T00:00:00"}