{"id":"https://openalex.org/W7216215399","doi":"https://doi.org/10.48550/arxiv.2610.01864","title":"From Isolated Feature to Orbits: Discovering Music Concepts via Multi-SAE Alignment","display_name":"From Isolated Feature to Orbits: Discovering Music Concepts via Multi-SAE Alignment","publication_year":2026,"publication_date":"2026-10-01","ids":{"openalex":"https://openalex.org/W7216215399","doi":"https://doi.org/10.48550/arxiv.2610.01864"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.01864","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.01864","pdf_url":"https://arxiv.org/pdf/2610.01864","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2610.01864","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5153451246","display_name":"Liwei Lin","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I258800397","display_name":"New York University Shanghai","ror":"https://ror.org/02vpsdb40","country_code":"CN","type":"education","lineage":["https://openalex.org/I258800397","https://openalex.org/I57206974"]},{"id":"https://openalex.org/I4210113480","display_name":"Mohamed bin Zayed University of Artificial Intelligence","ror":"https://ror.org/0258gkt32","country_code":"AE","type":"education","lineage":["https://openalex.org/I4210113480"]}],"countries":["AE","CN"],"is_corresponding":false,"raw_author_name":"Lin, Liwei","raw_affiliation_strings":["Computer Science Department , NYU Shanghai Machine Learning Department , MBZUAI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Computer Science Department , NYU Shanghai Machine Learning Department , MBZUAI","institution_ids":["https://openalex.org/I258800397","https://openalex.org/I4210113480"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5155023247","display_name":"Gus Xia","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4210113480","display_name":"Mohamed bin Zayed University of Artificial Intelligence","ror":"https://ror.org/0258gkt32","country_code":"AE","type":"education","lineage":["https://openalex.org/I4210113480"]}],"countries":["AE"],"is_corresponding":false,"raw_author_name":"Xia, Gus","raw_affiliation_strings":["Machine Learning Department , MBZUAI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Machine Learning Department , MBZUAI","institution_ids":["https://openalex.org/I4210113480"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.7773000001907349,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.7773000001907349,"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/T11309","display_name":"Music and Audio Processing","score":0.19589999318122864,"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/T11349","display_name":"Music Technology and Sound Studies","score":0.016599999740719795,"subfield":{"id":"https://openalex.org/subfields/1210","display_name":"Music"},"field":{"id":"https://openalex.org/fields/12","display_name":"Arts and Humanities"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/sparse-autoencoders","display_name":"sparse autoencoders","score":0.9390000104904175},{"id":"https://openalex.org/keywords/multi-view-alignment","display_name":"multi-view alignment","score":0.5619999766349792},{"id":"https://openalex.org/keywords/interpretability","display_name":"interpretability","score":0.5849999785423279},{"id":"https://openalex.org/keywords/orbit","display_name":"orbit","score":0.4790000021457672},{"id":"https://openalex.org/keywords/chords","display_name":"chords","score":0.765999972820282},{"id":"https://openalex.org/keywords/musical-key","display_name":"musical key","score":0.859000027179718},{"id":"https://openalex.org/keywords/melodic-patterns","display_name":"melodic patterns","score":0.8489999771118164},{"id":"https://openalex.org/keywords/inductive-bias","display_name":"inductive bias","score":0.7089999914169312},{"id":"https://openalex.org/keywords/music-representation-learning","display_name":"music representation learning","score":0.5580000281333923}],"concepts":[{"id":"https://openalex.org/C194147245","wikidata":"https://www.wikidata.org/wiki/Q1076368","display_name":"Chord (peer-to-peer)","level":2,"score":0.8755000233650208},{"id":"https://openalex.org/C2781067378","wikidata":"https://www.wikidata.org/wiki/Q17027399","display_name":"Interpretability","level":2,"score":0.8015000224113464},{"id":"https://openalex.org/C43803900","wikidata":"https://www.wikidata.org/wiki/Q170412","display_name":"Melody","level":3,"score":0.6593999862670898},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.6291000247001648},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6083999872207642},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.5875999927520752},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5401999950408936},{"id":"https://openalex.org/C2780966255","wikidata":"https://www.wikidata.org/wiki/Q5474306","display_name":"Foundation (evidence)","level":2,"score":0.4577000141143799},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.37790000438690186},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3700000047683716},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.36169999837875366},{"id":"https://openalex.org/C12455157","wikidata":"https://www.wikidata.org/wiki/Q7835331","display_name":"Transposition (logic)","level":2,"score":0.34529998898506165},{"id":"https://openalex.org/C2777946086","wikidata":"https://www.wikidata.org/wiki/Q1163335","display_name":"Music information retrieval","level":3,"score":0.3212999999523163},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.30640000104904175},{"id":"https://openalex.org/C83665646","wikidata":"https://www.wikidata.org/wiki/Q42139305","display_name":"Feature vector","level":2,"score":0.3052000105381012},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.2955999970436096},{"id":"https://openalex.org/C143857728","wikidata":"https://www.wikidata.org/wiki/Q193544","display_name":"Music theory","level":3,"score":0.2946999967098236},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.29159998893737793},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2913999855518341},{"id":"https://openalex.org/C2778744346","wikidata":"https://www.wikidata.org/wiki/Q1152224","display_name":"Distinctive feature","level":2,"score":0.2712000012397766},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.25040000677108765}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.01864","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.01864","pdf_url":"https://arxiv.org/pdf/2610.01864","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2610.01864","doi":"https://doi.org/10.48550/arxiv.2610.01864","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.01864","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2610.01864","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.01864","pdf_url":"https://arxiv.org/pdf/2610.01864","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7216215399.pdf","grobid_xml":"https://content.openalex.org/works/W7216215399.grobid-xml"},"referenced_works_count":4,"referenced_works":["https://openalex.org/W1853900790","https://openalex.org/W4386839891","https://openalex.org/W7128021905","https://openalex.org/W7140093518"],"related_works":["https://openalex.org/W7172335607","https://openalex.org/W7203708261","https://openalex.org/W4414580999","https://openalex.org/W3207313418","https://openalex.org/W6949966000","https://openalex.org/W7116884112","https://openalex.org/W2747811593","https://openalex.org/W6949660601","https://openalex.org/W7172259714","https://openalex.org/W7168156614"],"abstract_inverted_index":{"How":[0],"can":[1],"we":[2,126,147],"understand":[3],"what":[4],"a":[5,83,90,128,192],"music":[6,114,183],"foundation":[7,115,184],"model":[8,116],"has":[9],"learned":[10,110],"\\textit{internally}?":[11],"Most":[12],"interpretability":[13],"approaches,":[14],"such":[15,67],"as":[16,39,68,75,118,134],"probing":[17],"and":[18,62,152,177],"Sparse":[19],"Autoencoders":[20],"(SAEs),":[21],"focus":[22],"on":[23],"identifying":[24],"individual":[25],"features":[26],"with":[27],"minimal":[28,189],"structural":[29],"assumptions.":[30],"We":[31],"argue":[32],"that":[33,130,167],"many":[34],"concepts":[35,66],"are":[36,55,72],"better":[37],"understood":[38],"\\textit{structured":[40],"relations}":[41],"rather":[42,92],"than":[43,93],"isolated":[44,94],"features.":[45,95,122,163],"This":[46],"is":[47],"especially":[48],"prominent":[49],"in":[50,57],"music,":[51],"where":[52],"tonal":[53],"structures":[54,120,172],"organized":[56,119],"the":[58,79,86,109],"space":[59],"of":[60,82,113,161],"pitch":[61,132],"time.":[63],"For":[64],"example,":[65],"chords":[69],"or":[70,85],"keys":[71],"naturally":[73],"expressed":[74],"structured":[76,159],"sets":[77],"(e.g.,":[78,191],"12":[80],"transpositions":[81],"chord":[84],"diatonic":[87],"system":[88],"within":[89],"key),":[91],"In":[96],"this":[97,124,168],"study,":[98],"\\textbf{we":[99],"shift":[100],"from":[101],"feature":[102],"identification":[103],"to":[104,138,157,174,196],"structure-based":[105],"analysis},":[106],"asking":[107],"whether":[108],"inner":[111],"representations":[112,156],"emerge":[117],"over":[121],"To":[123],"end,":[125],"introduce":[127],"framework":[129],"uses":[131],"transposition":[133],"an":[135],"inductive":[136],"bias":[137],"induce":[139],"ordered":[140],"orbits":[141],"via":[142],"multi-view":[143],"SAE":[144,155],"alignment.":[145],"Concretely,":[146],"generate":[148],"pitch-shifted":[149],"input":[150],"pairs":[151],"align":[153],"their":[154],"discover":[158],"groups":[160],"pitch-related":[162],"Experimental":[164],"results":[165],"show":[166],"approach":[169],"recovers":[170],"orbit":[171],"corresponding":[173],"chords,":[175],"keys,":[176],"melodic":[178],"patterns":[179],"across":[180],"two":[181],"state-of-the-art":[182],"models,":[185],"while":[186],"requiring":[187],"only":[188],"grounding":[190],"few":[193],"anchor":[194],"examples)":[195],"interpret":[197],"entire":[198],"concept":[199],"families.":[200]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-10-03T00:00:00"}