{"id":"https://openalex.org/W7166830504","doi":"https://doi.org/10.18653/v1/2026.findings-acl.537","title":"[b] = [d] - [t] + [p]: Self-supervised Speech Models Discover Phonological Vector Arithmetic","display_name":"[b] = [d] - [t] + [p]: Self-supervised Speech Models Discover Phonological Vector Arithmetic","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166830504","doi":"https://doi.org/10.18653/v1/2026.findings-acl.537"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.537","doi":"https://doi.org/10.18653/v1/2026.findings-acl.537","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.537","pdf_url":"https://aclanthology.org/2026.findings-acl.537.pdf","source":{"id":"https://openalex.org/S4363605144","display_name":"Findings of the Association for Computational Linguistics: ACL (Findings of ACL)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://aclanthology.org/2026.findings-acl.537.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139827127","display_name":"Kwanghee Choi","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kwanghee Choi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110721529","display_name":"Eunjung Yeo","orcid":"https://orcid.org/0009-0003-4452-4084","observed_orcids":["https://orcid.org/0009-0003-4452-4084"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Eunjung Yeo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015390421","display_name":"Cheol Jun Cho","orcid":"https://orcid.org/0000-0003-2596-3164","observed_orcids":["https://orcid.org/0000-0003-2596-3164"]},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Cheol Jun Cho","raw_affiliation_strings":["UC Berkeley ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Berkeley ,","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5004717608","display_name":"David F. Harwath","orcid":"https://orcid.org/0000-0003-0206-0253","observed_orcids":["https://orcid.org/0000-0003-0206-0253"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"David Harwath","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5059859009","display_name":"David R. Mortensen","orcid":"https://orcid.org/0000-0002-3927-6851","observed_orcids":["https://orcid.org/0000-0002-3927-6851"]},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"David R. Mortensen","raw_affiliation_strings":["CMU"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"CMU","institution_ids":["https://openalex.org/I74973139"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.45399503,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"11048","last_page":"11069"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.8012999892234802,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.8012999892234802,"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/T10403","display_name":"Phonetics and Phonology Research","score":0.1843000054359436,"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":0.0035000001080334187,"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/self-supervised-speech-models","display_name":"self-supervised speech models","score":0.9800000190734863},{"id":"https://openalex.org/keywords/phonological-features","display_name":"phonological features","score":0.7409999966621399},{"id":"https://openalex.org/keywords/phonetic-representation","display_name":"phonetic representation","score":0.6100000143051147},{"id":"https://openalex.org/keywords/voicing","display_name":"voicing","score":0.6579999923706055},{"id":"https://openalex.org/keywords/compositional-representations","display_name":"compositional representations","score":0.5809999704360962},{"id":"https://openalex.org/keywords/cross-linguistic-phonetics","display_name":"cross-linguistic phonetics","score":0.6990000009536743}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5879999995231628},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.49639999866485596},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46720001101493835},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.37720000743865967},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.29739999771118164},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.2791999876499176},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.26980000734329224},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.266400009393692},{"id":"https://openalex.org/C94375191","wikidata":"https://www.wikidata.org/wiki/Q11205","display_name":"Arithmetic","level":1,"score":0.26499998569488525},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.263700008392334},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.26159998774528503}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.537","doi":"https://doi.org/10.18653/v1/2026.findings-acl.537","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.537","pdf_url":"https://aclanthology.org/2026.findings-acl.537.pdf","source":{"id":"https://openalex.org/S4363605144","display_name":"Findings of the Association for Computational Linguistics: ACL (Findings of ACL)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.537","doi":"https://doi.org/10.18653/v1/2026.findings-acl.537","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.537","pdf_url":"https://aclanthology.org/2026.findings-acl.537.pdf","source":{"id":"https://openalex.org/S4363605144","display_name":"Findings of the Association for Computational Linguistics: ACL (Findings of ACL)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G4732812235","display_name":"CAREER: Modeling Spoken Language Without Parallel Text Annotations","funder_award_id":"2238605","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166830504.pdf","grobid_xml":"https://content.openalex.org/works/W7166830504.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7136730862","https://openalex.org/W7083808329","https://openalex.org/W4402112198","https://openalex.org/W4385822337","https://openalex.org/W4417356807","https://openalex.org/W4394773771","https://openalex.org/W7168184922","https://openalex.org/W7162453962","https://openalex.org/W4415433975","https://openalex.org/W7161328609"],"abstract_inverted_index":{"Self-supervised":[0],"speech":[1,115],"models":[2],"(S3Ms)":[3],"are":[4,130],"known":[5],"to":[6,26,37,54,67,96],"encode":[7,114],"rich":[8],"phonetic":[9],"information,":[10],"yet":[11],"how":[12],"this":[13,94],"information":[14],"is":[15],"structured":[16],"remains":[17],"underexplored.We":[18],"conduct":[19],"a":[20,79,90,105],"comprehensive":[21],"study":[22],"across":[23],"96":[24],"languages":[25],"analyze":[27],"the":[28,48,60,68,83],"underlying":[29],"structure":[30],"of":[31,62,70,73,107],"S3M":[32],"representations,":[33],"with":[34],"particular":[35],"attention":[36],"phonological":[38,55,64,76,123],"vectors.We":[39],"first":[40],"show":[41],"that":[42,52,59,112],"there":[43],"exist":[44],"linear":[45],"directions":[46],"within":[47],"model's":[49],"representation":[50],"space":[51],"correspond":[53],"features.We":[56],"further":[57],"demonstrate":[58],"scale":[61],"these":[63,109],"vectors":[65],"correlate":[66],"degree":[69],"acoustic":[71],"realization":[72],"their":[74],"corresponding":[75],"features":[77],"in":[78,104],"continuous":[80],"manner.For":[81],"example,":[82],"difference":[84],"between":[85],"[d]":[86],"and":[87,119,127],"[t]":[88],"yields":[89],"voicing":[91],"vector:":[92],"adding":[93],"vector":[95,124],"[p]":[97],"produces":[98],"[b],":[99],"while":[100],"scaling":[101],"it":[102],"results":[103],"continuum":[106],"voicing.Together,":[108],"findings":[110],"indicate":[111],"S3Ms":[113],"using":[116],"phonologically":[117],"interpretable":[118],"compositional":[120],"vectors,":[121],"demonstrating":[122],"arithmetic.All":[125],"code":[126],"interactive":[128],"demos":[129],"available":[131],"at":[132],"https://github.com/":[133],"juice500ml/phonetic-arithmetic.":[134]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-07-02T00:00:00"}