{"id":"https://openalex.org/W4408353405","doi":"https://doi.org/10.1109/icassp49660.2025.10888472","title":"Rethinking Decoding in Multi-intent Spoken Language Understanding","display_name":"Rethinking Decoding in Multi-intent Spoken Language Understanding","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408353405","doi":"https://doi.org/10.1109/icassp49660.2025.10888472"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10888472","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10888472","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5062996827","display_name":"Ying Xia","orcid":"https://orcid.org/0000-0002-7407-6126"},"institutions":[{"id":"https://openalex.org/I113940042","display_name":"Shanghai University","ror":"https://ror.org/006teas31","country_code":"CN","type":"education","lineage":["https://openalex.org/I113940042"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ying Xia","raw_affiliation_strings":["Shanghai University,School of Computer Engineering and Science"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai University,School of Computer Engineering and Science","institution_ids":["https://openalex.org/I113940042"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100546007","display_name":"Zhen Xiong","orcid":null},"institutions":[{"id":"https://openalex.org/I75390827","display_name":"Beijing University of Chemical Technology","ror":"https://ror.org/00df5yc52","country_code":"CN","type":"education","lineage":["https://openalex.org/I75390827"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhen Xiong","raw_affiliation_strings":["Beijing University of Chemical Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Beijing University of Chemical Technology","institution_ids":["https://openalex.org/I75390827"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5037026475","display_name":"Kefan Shen","orcid":null},"institutions":[{"id":"https://openalex.org/I113940042","display_name":"Shanghai University","ror":"https://ror.org/006teas31","country_code":"CN","type":"education","lineage":["https://openalex.org/I113940042"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Kefan Shen","raw_affiliation_strings":["Shanghai University,School of Computer Engineering and Science"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai University,School of Computer Engineering and Science","institution_ids":["https://openalex.org/I113940042"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101006722","display_name":"Zhihong Zhu","orcid":null},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhihong Zhu","raw_affiliation_strings":["Tencent YouTu Lab"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tencent YouTu Lab","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5086702173","display_name":"Shaorong Xie","orcid":"https://orcid.org/0000-0002-8016-9310"},"institutions":[{"id":"https://openalex.org/I113940042","display_name":"Shanghai University","ror":"https://ror.org/006teas31","country_code":"CN","type":"education","lineage":["https://openalex.org/I113940042"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shaorong Xie","raw_affiliation_strings":["Shanghai University,School of Computer Engineering and Science"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai University,School of Computer Engineering and Science","institution_ids":["https://openalex.org/I113940042"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100641142","display_name":"Wei Liu","orcid":"https://orcid.org/0000-0002-7409-0948"},"institutions":[{"id":"https://openalex.org/I113940042","display_name":"Shanghai University","ror":"https://ror.org/006teas31","country_code":"CN","type":"education","lineage":["https://openalex.org/I113940042"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wei Liu","raw_affiliation_strings":["Shanghai University,School of Computer Engineering and Science"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai University,School of Computer Engineering and Science","institution_ids":["https://openalex.org/I113940042"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.673,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.79080814,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12031","display_name":"Speech and dialogue systems","score":0.9330000281333923,"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/T12031","display_name":"Speech and dialogue systems","score":0.9330000281333923,"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/decoding-methods","display_name":"Decoding methods","score":0.7910813689231873},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7202708721160889},{"id":"https://openalex.org/keywords/spoken-language","display_name":"Spoken language","score":0.4982724189758301},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.38461482524871826},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.3414851427078247},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.3229252099990845},{"id":"https://openalex.org/keywords/telecommunications","display_name":"Telecommunications","score":0.13561606407165527}],"concepts":[{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.7910813689231873},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7202708721160889},{"id":"https://openalex.org/C2776230583","wikidata":"https://www.wikidata.org/wiki/Q1322198","display_name":"Spoken language","level":2,"score":0.4982724189758301},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.38461482524871826},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.3414851427078247},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3229252099990845},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.13561606407165527},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49660.2025.10888472","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10888472","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.7200000286102295,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":30,"referenced_works":["https://openalex.org/W2077302143","https://openalex.org/W2803392141","https://openalex.org/W2946085385","https://openalex.org/W3098800734","https://openalex.org/W3161463965","https://openalex.org/W3173794693","https://openalex.org/W3188188945","https://openalex.org/W3194955531","https://openalex.org/W4283218202","https://openalex.org/W4312303628","https://openalex.org/W4362698731","https://openalex.org/W4375868795","https://openalex.org/W4382119366","https://openalex.org/W4382202608","https://openalex.org/W4385572550","https://openalex.org/W4385573163","https://openalex.org/W4385573420","https://openalex.org/W4388979610","https://openalex.org/W4392366643","https://openalex.org/W4392909978","https://openalex.org/W4393157386","https://openalex.org/W4401024992","https://openalex.org/W4408355719","https://openalex.org/W6631190155","https://openalex.org/W6751425476","https://openalex.org/W6755207826","https://openalex.org/W6759393636","https://openalex.org/W6795453751","https://openalex.org/W6843814252","https://openalex.org/W6856029324"],"related_works":["https://openalex.org/W2161474341","https://openalex.org/W4302615923","https://openalex.org/W3203142394","https://openalex.org/W2351061015","https://openalex.org/W4220731478","https://openalex.org/W1974101135","https://openalex.org/W2017509870","https://openalex.org/W4251141768","https://openalex.org/W4360952157","https://openalex.org/W633922780"],"abstract_inverted_index":{"Multi-intent":[0],"spoken":[1],"language":[2],"understanding":[3],"(SLU)":[4],"can":[5],"handle":[6],"multiple":[7],"intent":[8,33,101],"utterances":[9],"in":[10,36,121],"real-world":[11],"scenarios,":[12],"which":[13,61],"has":[14],"gained":[15],"increasing":[16],"research":[17],"attention.":[18],"Despite":[19],"promising":[20],"results":[21,135],"achieved":[22],"by":[23],"existing":[24],"joint":[25],"models,":[26],"they":[27],"(1)":[28],"perform":[29],"utterance-level":[30],"or":[31,43],"token-level":[32],"detection,":[34],"resulting":[35],"suboptimal":[37],"performance":[38],"due":[39],"to":[40,81,98,125],"fixed":[41],"thresholds":[42],"voting":[44],"mechanisms;":[45],"(2)":[46],"incorporate":[47],"all":[48],"predicted":[49,110],"intents":[50,111],"into":[51],"each":[52],"slot":[53,59,71,132],"hidden":[54],"state":[55],"and":[56,66,116],"execute":[57],"parallel":[58],"decoding,":[60],"lacks":[62],"precise":[63,113,131],"intent-slot":[64,114],"alignment":[65],"overlooks":[67],"sequential":[68,128],"dependencies":[69,129],"between":[70],"tokens.":[72],"In":[73],"this":[74],"paper,":[75],"we":[76,90,107],"propose":[77],"a":[78,92],"new":[79],"framework":[80],"tackle":[82],"these":[83],"two":[84,137],"issues.":[85],"For":[86,103],"the":[87,104,122,140],"first":[88],"issue,":[89,106],"utilize":[91],"global":[93],"pointer":[94],"with":[95],"auxiliary":[96],"tasks":[97],"achieve":[99],"span-based":[100,109],"detection.":[102],"second":[105],"leverage":[108],"for":[112,130],"guidance":[115],"introduce":[117],"rotational":[118],"position":[119],"encoding":[120],"interaction":[123],"module":[124],"explicitly":[126],"model":[127],"filling.":[133],"Experimental":[134],"on":[136],"benchmarks":[138],"demonstrate":[139],"superiority":[141],"of":[142],"our":[143],"framework.":[144]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}