{"id":"https://openalex.org/W7208767237","doi":"https://doi.org/10.18653/2026.findings-emnlp.1400","title":"Listen to the Latents: Self-Correcting Speech Recognition in Large Audio Language Models Through Hidden-State Interactions","display_name":"Listen to the Latents: Self-Correcting Speech Recognition in Large Audio Language Models Through Hidden-State Interactions","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7208767237","doi":"https://doi.org/10.18653/2026.findings-emnlp.1400"},"language":"en","primary_location":{"id":"doi:10.18653/2026.findings-emnlp.1400","doi":"https://doi.org/10.18653/2026.findings-emnlp.1400","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.findings-emnlp.1400","pdf_url":"https://aclanthology.org/2026.findings-emnlp.1400.pdf","source":{"id":"https://openalex.org/S7407096085","display_name":"Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"conference"},"license":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2026","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://aclanthology.org/2026.findings-emnlp.1400.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5059902893","display_name":"Chan-Jan Hsu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I16733864","display_name":"National Taiwan University","ror":"https://ror.org/05bqach95","country_code":"TW","type":"education","lineage":["https://openalex.org/I16733864"]},{"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":["TW","US"],"is_corresponding":false,"raw_author_name":"Chan-Jan Hsu","raw_affiliation_strings":["Carnegie Mellon University \u266d National Taiwan University \u266e"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University \u266d National Taiwan University \u266e","institution_ids":["https://openalex.org/I16733864","https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139247149","display_name":"Jaeyeon Kim","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I16733864","display_name":"National Taiwan University","ror":"https://ror.org/05bqach95","country_code":"TW","type":"education","lineage":["https://openalex.org/I16733864"]},{"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":["TW","US"],"is_corresponding":false,"raw_author_name":"Jaeyeon Kim","raw_affiliation_strings":["Carnegie Mellon University \u266d National Taiwan University \u266e"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University \u266d National Taiwan University \u266e","institution_ids":["https://openalex.org/I16733864","https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5123494971","display_name":"Chao-Han Huck Yang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chao-Han Huck Yang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059482281","display_name":"Shinji Watanabe","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I16733864","display_name":"National Taiwan University","ror":"https://ror.org/05bqach95","country_code":"TW","type":"education","lineage":["https://openalex.org/I16733864"]},{"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":["TW","US"],"is_corresponding":false,"raw_author_name":"Shinji Watanabe","raw_affiliation_strings":["Carnegie Mellon University \u266d National Taiwan University \u266e"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University \u266d National Taiwan University \u266e","institution_ids":["https://openalex.org/I16733864","https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139212885","display_name":"Hung-yi Lee","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I16733864","display_name":"National Taiwan University","ror":"https://ror.org/05bqach95","country_code":"TW","type":"education","lineage":["https://openalex.org/I16733864"]},{"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":["TW","US"],"is_corresponding":false,"raw_author_name":"Hung-yi Lee","raw_affiliation_strings":["Carnegie Mellon University \u266d National Taiwan University \u266e"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University \u266d National Taiwan University \u266e","institution_ids":["https://openalex.org/I16733864","https://openalex.org/I74973139"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5123953982","display_name":"Carlos Busso","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I1304085615","display_name":"Nvidia (United Kingdom)","ror":"https://ror.org/02kr42612","country_code":"GB","type":"company","lineage":["https://openalex.org/I1304085615","https://openalex.org/I4210127875"]},{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Carlos Busso","raw_affiliation_strings":["NVIDIA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NVIDIA","institution_ids":["https://openalex.org/I1304085615","https://openalex.org/I4210127875"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":4,"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":0.70774334,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"29818","last_page":"29841"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":1.0,"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":1.0,"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/T10860","display_name":"Speech and Audio Processing","score":0.0,"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/T10028","display_name":"Topic Modeling","score":0.0,"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/large-audio-language-models","display_name":"large audio language models","score":0.6809999942779541},{"id":"https://openalex.org/keywords/lora-adaptation","display_name":"LoRA adaptation","score":0.5809999704360962},{"id":"https://openalex.org/keywords/hybrid-search","display_name":"hybrid retrieval","score":0.6510000228881836}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8079000115394592},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.7684000134468079},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.633400022983551},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5192999839782715},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5105000138282776},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4733000099658966},{"id":"https://openalex.org/C4554734","wikidata":"https://www.wikidata.org/wiki/Q593744","display_name":"Knowledge base","level":2,"score":0.35499998927116394},{"id":"https://openalex.org/C23224414","wikidata":"https://www.wikidata.org/wiki/Q176769","display_name":"Hidden Markov model","level":2,"score":0.33809998631477356},{"id":"https://openalex.org/C2776230583","wikidata":"https://www.wikidata.org/wiki/Q1322198","display_name":"Spoken language","level":2,"score":0.3345000147819519},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.3149999976158142},{"id":"https://openalex.org/C42058472","wikidata":"https://www.wikidata.org/wiki/Q810214","display_name":"Base (topology)","level":2,"score":0.29840001463890076},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.29589998722076416},{"id":"https://openalex.org/C155635449","wikidata":"https://www.wikidata.org/wiki/Q4674699","display_name":"Acoustic model","level":3,"score":0.2678000032901764},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.26089999079704285},{"id":"https://openalex.org/C2983448237","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Language understanding","level":2,"score":0.25929999351501465},{"id":"https://openalex.org/C54953205","wikidata":"https://www.wikidata.org/wiki/Q4142201","display_name":"Speech analytics","level":4,"score":0.25369998812675476}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/2026.findings-emnlp.1400","doi":"https://doi.org/10.18653/2026.findings-emnlp.1400","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.findings-emnlp.1400","pdf_url":"https://aclanthology.org/2026.findings-emnlp.1400.pdf","source":{"id":"https://openalex.org/S7407096085","display_name":"Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"conference"},"license":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2026","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2609.02940","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.02940","pdf_url":"https://arxiv.org/pdf/2609.02940","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2609.02940","doi":"https://doi.org/10.48550/arxiv.2609.02940","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.02940","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":"doi:10.18653/2026.findings-emnlp.1400","doi":"https://doi.org/10.18653/2026.findings-emnlp.1400","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.findings-emnlp.1400","pdf_url":"https://aclanthology.org/2026.findings-emnlp.1400.pdf","source":{"id":"https://openalex.org/S7407096085","display_name":"Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"conference"},"license":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7208767237.pdf","grobid_xml":"https://content.openalex.org/works/W7208767237.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4402112487","https://openalex.org/W4387800213","https://openalex.org/W4399062157","https://openalex.org/W7133205981","https://openalex.org/W7151422764","https://openalex.org/W7048003894","https://openalex.org/W7213373494","https://openalex.org/W4415250109","https://openalex.org/W4386907919","https://openalex.org/W4411249889"],"abstract_inverted_index":{"Recent":[0],"automatic":[1],"speech":[2],"recognition":[3],"(ASR)":[4],"systems":[5],"increasingly":[6],"integrate":[7],"large":[8],"language":[9],"models":[10,43,135],"(LLMs)":[11],"to":[12,28,142],"leverage":[13,138],"their":[14,46,139],"semantic":[15,98,106,127],"knowledge,":[16],"either":[17],"externally":[18],"through":[19,24,130],"logit":[20],"fusion":[21],"or":[22],"internally":[23],"warm":[25,131],"initialization.However,":[26],"how":[27],"effectively":[29],"combine":[30],"these":[31],"two":[32,73],"strategies":[33],"remains":[34],"underexplored.In":[35],"this":[36],"work,":[37],"we":[38,63],"refine":[39],"warm-initialized":[40],"LLM-based":[41,82,133],"ASR":[42,83,109,134],"by":[44,72],"leveraging":[45],"own":[47],"preadaptation":[48],"base":[49,57,140],"LLMs,":[50],"focusing":[51],"on":[52],"LoRAadapted":[53],"settings":[54],"where":[55],"the":[56,79],"LLM":[58,141],"is":[59],"preserved.To":[60],"achieve":[61],"this,":[62],"propose":[64],"Hybrid":[65],"Search,":[66],"a":[67,94],"targeted":[68,102],"correction":[69],"strategy":[70],"motivated":[71],"observations.First,":[74],"interaction":[75],"features":[76],"that":[77],"characterize":[78],"relationship":[80],"between":[81],"hidden":[84,88],"states":[85,89],"and":[86,119],"base-LLM":[87],"provide":[90],"informative":[91],"signals":[92],"about":[93],"token's":[95],"degree":[96],"of":[97],"dependence.Second,":[99],"selectively":[100],"refining":[101],"tokens":[103],"with":[104],"high":[105],"dependence":[107],"improves":[108],"performance":[110],"far":[111],"beyond":[112],"naive":[113],"global":[114],"LLM-correction":[115],"methods":[116],"including":[117],"rescoring":[118],"late":[120],"fusion.Our":[121],"analysis":[122],"suggests":[123],"that,":[124],"even":[125],"after":[126],"knowledge":[128],"transfer":[129],"initialization,":[132],"can":[136],"still":[137],"further":[143],"improve":[144],"inference-time":[145],"performance.":[146]},"counts_by_year":[],"updated_date":"2026-11-01T11:40:01.658920","created_date":"2026-09-05T00:00:00"}