{"id":"https://openalex.org/W4415931574","doi":"https://doi.org/10.1145/3746252.3761405","title":"As Good as It KAN Get: High-Fidelity Audio Representation","display_name":"As Good as It KAN Get: High-Fidelity Audio Representation","publication_year":2025,"publication_date":"2025-11-08","ids":{"openalex":"https://openalex.org/W4415931574","doi":"https://doi.org/10.1145/3746252.3761405"},"language":"en","primary_location":{"id":"doi:10.1145/3746252.3761405","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746252.3761405","pdf_url":null,"source":null,"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 34th ACM International Conference on Information and Knowledge Management","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3746252.3761405","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5120269598","display_name":"Patryk Marsza\u0142ek","orcid":"https://orcid.org/0009-0008-7695-7905"},"institutions":[{"id":"https://openalex.org/I126596746","display_name":"Jagiellonian University","ror":"https://ror.org/03bqmcz70","country_code":"PL","type":"education","lineage":["https://openalex.org/I126596746"]}],"countries":["PL"],"is_corresponding":false,"raw_author_name":"Patryk Marsza\u0142ek","raw_affiliation_strings":["Jagiellonian University, Krakow, Poland"],"raw_orcid":"https://orcid.org/0009-0008-7695-7905","affiliations":[{"raw_affiliation_string":"Jagiellonian University, Krakow, Poland","institution_ids":["https://openalex.org/I126596746"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120269599","display_name":"Maciej Rut","orcid":"https://orcid.org/0009-0009-9637-6128"},"institutions":[{"id":"https://openalex.org/I126596746","display_name":"Jagiellonian University","ror":"https://ror.org/03bqmcz70","country_code":"PL","type":"education","lineage":["https://openalex.org/I126596746"]}],"countries":["PL"],"is_corresponding":false,"raw_author_name":"Maciej Rut","raw_affiliation_strings":["Jagiellonian University, Krakow, Poland"],"raw_orcid":"https://orcid.org/0009-0009-9637-6128","affiliations":[{"raw_affiliation_string":"Jagiellonian University, Krakow, Poland","institution_ids":["https://openalex.org/I126596746"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032526628","display_name":"Piotr Kawa","orcid":"https://orcid.org/0000-0002-2025-0547"},"institutions":[{"id":"https://openalex.org/I11923345","display_name":"Wroc\u0142aw University of Science and Technology","ror":"https://ror.org/008fyn775","country_code":"PL","type":"education","lineage":["https://openalex.org/I11923345"]},{"id":"https://openalex.org/I686019","display_name":"AGH University of Krakow","ror":"https://ror.org/00bas1c41","country_code":"PL","type":"education","lineage":["https://openalex.org/I686019"]}],"countries":["PL"],"is_corresponding":false,"raw_author_name":"Piotr Kawa","raw_affiliation_strings":["Wroclaw University of Science and Technology, Wroclaw, Poland"],"raw_orcid":"https://orcid.org/0000-0002-2025-0547","affiliations":[{"raw_affiliation_string":"Wroclaw University of Science and Technology, Wroclaw, Poland","institution_ids":["https://openalex.org/I11923345","https://openalex.org/I686019"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5068511223","display_name":"Przemys\u0142aw Spurek","orcid":"https://orcid.org/0000-0003-0097-5521"},"institutions":[{"id":"https://openalex.org/I126596746","display_name":"Jagiellonian University","ror":"https://ror.org/03bqmcz70","country_code":"PL","type":"education","lineage":["https://openalex.org/I126596746"]}],"countries":["PL"],"is_corresponding":false,"raw_author_name":"Przemys\u0142aw Spurek","raw_affiliation_strings":["Jagiellonian University, Krakow, Poland and IDEAS Research Institute, Warszawa, Poland"],"raw_orcid":"https://orcid.org/0000-0003-0097-5521","affiliations":[{"raw_affiliation_string":"Jagiellonian University, Krakow, Poland and IDEAS Research Institute, Warszawa, Poland","institution_ids":["https://openalex.org/I126596746"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5088028973","display_name":"Piotr Syga","orcid":"https://orcid.org/0000-0002-0266-5802"},"institutions":[{"id":"https://openalex.org/I11923345","display_name":"Wroc\u0142aw University of Science and Technology","ror":"https://ror.org/008fyn775","country_code":"PL","type":"education","lineage":["https://openalex.org/I11923345"]},{"id":"https://openalex.org/I686019","display_name":"AGH University of Krakow","ror":"https://ror.org/00bas1c41","country_code":"PL","type":"education","lineage":["https://openalex.org/I686019"]}],"countries":["PL"],"is_corresponding":false,"raw_author_name":"Piotr Syga","raw_affiliation_strings":["Wroclaw University of Science and Technology, Wroclaw, Poland"],"raw_orcid":"https://orcid.org/0000-0002-0266-5802","affiliations":[{"raw_affiliation_string":"Wroclaw University of Science and Technology, Wroclaw, Poland","institution_ids":["https://openalex.org/I11923345","https://openalex.org/I686019"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":2.0842,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":{"value":0.89206526,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":98,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"2063","last_page":"2073"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.7560999989509583,"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/T10860","display_name":"Speech and Audio Processing","score":0.7560999989509583,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.1088000014424324,"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":0.06769999861717224,"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/representation","display_name":"Representation (politics)","score":0.6818000078201294},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.656499981880188},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.6315000057220459},{"id":"https://openalex.org/keywords/sound-quality","display_name":"Sound quality","score":0.5652999877929688},{"id":"https://openalex.org/keywords/perception","display_name":"Perception","score":0.510699987411499},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4634000062942505}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7652999758720398},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.6818000078201294},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.656499981880188},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6326000094413757},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.6315000057220459},{"id":"https://openalex.org/C167310288","wikidata":"https://www.wikidata.org/wiki/Q7564808","display_name":"Sound quality","level":2,"score":0.5652999877929688},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.510699987411499},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4634000062942505},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.4343999922275543},{"id":"https://openalex.org/C64922751","wikidata":"https://www.wikidata.org/wiki/Q4650799","display_name":"Audio signal","level":3,"score":0.4271000027656555},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.3953999876976013},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.39500001072883606},{"id":"https://openalex.org/C127220857","wikidata":"https://www.wikidata.org/wiki/Q2719318","display_name":"Audio signal processing","level":4,"score":0.353300005197525},{"id":"https://openalex.org/C13895895","wikidata":"https://www.wikidata.org/wiki/Q3270773","display_name":"Speech coding","level":2,"score":0.3336000144481659},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2964000105857849},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.27799999713897705}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1145/3746252.3761405","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746252.3761405","pdf_url":null,"source":null,"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 34th ACM International Conference on Information and Knowledge Management","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2503.02585","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2503.02585","pdf_url":"https://arxiv.org/pdf/2503.02585","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"pmh:oai:ruj.uj.edu.pl:item/565510","is_oa":true,"landing_page_url":"https://ruj.uj.edu.pl/handle/item/565510","pdf_url":"http://ruj.uj.edu.pl/bitstreams/fdb31b27-dc80-4c88-baa5-34270fcda998/download","source":{"id":"https://openalex.org/S4306401249","display_name":"Jagiellonian University Repository (Jagiellonian University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I126596746","host_organization_name":"Jagiellonian University","host_organization_lineage":["https://openalex.org/I126596746"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/conferenceObject"}],"best_oa_location":{"id":"doi:10.1145/3746252.3761405","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746252.3761405","pdf_url":null,"source":null,"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 34th ACM International Conference on Information and Knowledge Management","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320337041","display_name":"Department of Artificial Intelligence, Korea University","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Implicit":[0],"neural":[1],"representations":[2],"(INR)":[3],"have":[4],"gained":[5],"prominence":[6],"for":[7,39,67,114],"efficiently":[8],"encoding":[9],"multimedia":[10],"data,":[11],"yet":[12],"their":[13],"applications":[14],"in":[15,94,98],"audio":[16,40,109],"signals":[17],"remain":[18],"limited.":[19],"This":[20],"study":[21],"introduces":[22],"the":[23,51,58,87,112],"Kolmogorov-Arnold":[24],"Network":[25],"(KAN),":[26],"a":[27,77,91,105],"novel":[28],"architecture":[29,79],"using":[30],"learnable":[31],"activation":[32],"functions,":[33],"as":[34,104],"an":[35],"effective":[36],"INR":[37,82],"model":[38],"representation.":[41],"KAN":[42,103],"demonstrates":[43],"superior":[44],"perceptual":[45],"performance":[46],"over":[47],"previous":[48],"INRs,":[49],"achieving":[50],"lowest":[52],"Log-Spectral":[53],"Distance":[54],"of":[55,62,65],"1.29":[56],"and":[57,96,107,116],"highest":[59],"Perceptual":[60],"Evaluation":[61],"Speech":[63],"Quality":[64],"3.57":[66],"1.5~s":[68],"audio.":[69],"To":[70],"extend":[71],"KAN's":[72],"utility,":[73],"we":[74],"propose":[75],"FewSound,":[76],"hypernetwork-based":[78],"that":[80],"enhances":[81],"parameter":[83],"updates.":[84],"FewSound":[85],"outperforms":[86],"state-of-the-art":[88],"HyperSound,":[89],"with":[90,111],"33.3%":[92],"improvement":[93],"MSE":[95],"60.87%":[97],"SI-SNR.":[99],"These":[100],"results":[101],"show":[102],"robust":[106],"adaptable":[108],"representation":[110],"potential":[113],"scalability":[115],"integration":[117],"into":[118],"various":[119],"hypernetwork":[120],"frameworks.":[121]},"counts_by_year":[{"year":2026,"cited_by_count":2}],"updated_date":"2026-06-16T09:24:06.705377","created_date":"2025-11-05T00:00:00"}