{"id":"https://openalex.org/W4417445375","doi":"https://doi.org/10.18653/v1/2026.acl-long.693","title":"Protecting Bystander Privacy via Selective Hearing in Audio LLMs","display_name":"Protecting Bystander Privacy via Selective Hearing in Audio LLMs","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W4417445375","doi":"https://doi.org/10.18653/v1/2026.acl-long.693"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2026.acl-long.693","doi":"https://doi.org/10.18653/v1/2026.acl-long.693","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.693","pdf_url":null,"source":{"id":"https://openalex.org/S4393915740","display_name":"Proceedings of the conference - Association for Computational Linguistics. Meeting","issn_l":"0736-587X","issn":["0736-587X"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["norway-1"],"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":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","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://doi.org/10.18653/v1/2026.acl-long.693","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5097391710","display_name":"\u6653\u6167 \u6218","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiao Zhan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007755497","display_name":"Guangzhi Sun","orcid":"https://orcid.org/0000-0003-3190-497X","observed_orcids":["https://orcid.org/0000-0003-3190-497X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guangzhi Sun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Jose Such","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jose Such","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5002191410","display_name":"Philip C. Woodland","orcid":"https://orcid.org/0000-0001-9069-0225","observed_orcids":["https://orcid.org/0000-0001-9069-0225"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Phil Woodland","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.02500596,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"15180","last_page":"15192"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.6338000297546387,"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":0.6338000297546387,"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.18150000274181366,"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":0.07569999992847443,"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/large-audio-language-models","display_name":"large audio language models","score":0.8330000042915344},{"id":"https://openalex.org/keywords/privacy-leakage","display_name":"privacy leakage","score":0.7990000247955322},{"id":"https://openalex.org/keywords/privacy-preserving-fine-tuning","display_name":"privacy-preserving fine-tuning","score":0.593999981880188}],"concepts":[{"id":"https://openalex.org/C19106626","wikidata":"https://www.wikidata.org/wiki/Q230749","display_name":"Bystander effect","level":2,"score":0.9311000108718872},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5795000195503235},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5199000239372253},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5095000267028809},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.4975999891757965},{"id":"https://openalex.org/C2776889888","wikidata":"https://www.wikidata.org/wiki/Q1135789","display_name":"Unintended consequences","level":2,"score":0.4528999924659729},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.44679999351501465},{"id":"https://openalex.org/C511192102","wikidata":"https://www.wikidata.org/wiki/Q5156948","display_name":"Comprehension","level":2,"score":0.44679999351501465},{"id":"https://openalex.org/C108827166","wikidata":"https://www.wikidata.org/wiki/Q175975","display_name":"Internet privacy","level":1,"score":0.4268999993801117},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.39079999923706055},{"id":"https://openalex.org/C62230096","wikidata":"https://www.wikidata.org/wiki/Q275969","display_name":"Crowdsourcing","level":2,"score":0.29789999127388},{"id":"https://openalex.org/C95713431","wikidata":"https://www.wikidata.org/wiki/Q631425","display_name":"Vulnerability (computing)","level":2,"score":0.29760000109672546},{"id":"https://openalex.org/C2982805371","wikidata":"https://www.wikidata.org/wiki/Q104493","display_name":"Risk prevention","level":2,"score":0.2567000091075897}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/v1/2026.acl-long.693","doi":"https://doi.org/10.18653/v1/2026.acl-long.693","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.693","pdf_url":null,"source":{"id":"https://openalex.org/S4393915740","display_name":"Proceedings of the conference - Association for Computational Linguistics. Meeting","issn_l":"0736-587X","issn":["0736-587X"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["norway-1"],"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":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2512.06380","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2512.06380","pdf_url":"https://arxiv.org/pdf/2512.06380","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.2512.06380","doi":"https://doi.org/10.48550/arxiv.2512.06380","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2512.06380","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/v1/2026.acl-long.693","doi":"https://doi.org/10.18653/v1/2026.acl-long.693","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.693","pdf_url":null,"source":{"id":"https://openalex.org/S4393915740","display_name":"Proceedings of the conference - Association for Computational Linguistics. Meeting","issn_l":"0736-587X","issn":["0736-587X"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["norway-1"],"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":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7119557327","https://openalex.org/W7126054971","https://openalex.org/W7214941723","https://openalex.org/W4414734953","https://openalex.org/W7155054977","https://openalex.org/W4415230188","https://openalex.org/W7127110032","https://openalex.org/W7134992506","https://openalex.org/W4417229822","https://openalex.org/W7154069452"],"abstract_inverted_index":{"Audio":[0],"Large":[1],"language":[2],"models":[3,83,150],"(LLMs)":[4],"are":[5],"increasingly":[6],"deployed":[7],"in":[8,210],"the":[9,35,189,200],"real":[10],"world,":[11],"where":[12],"they":[13],"inevitably":[14],"capture":[15],"speech":[16],"from":[17],"unintended":[18],"nearby":[19],"bystanders,":[20],"raising":[21],"privacy":[22,118,209],"risks":[23],"that":[24,81,148,161],"existing":[25],"benchmarks":[26],"and":[27,73,86,104,112,176,197,206],"defences":[28],"did":[29],"not":[30],"consider.":[31],"We":[32,159],"introduce":[33],"SH-Bench,":[34],"first":[36,201],"benchmark":[37],"designed":[38],"to":[39,46,48,55,125,151,183],"evaluate":[40],"selective":[41,87,128,174],"hearing:":[42],"a":[43,97,144],"model's":[44],"ability":[45],"attend":[47],"an":[49,167,177],"intended":[50],"main":[51],"speaker":[52],"while":[53],"refusing":[54],"process":[56],"or":[57],"reveal":[58],"information":[59],"about":[60],"incidental":[61],"bystander":[62,117,131,171,208],"speech.":[63],"SH-Bench":[64,196],"contains":[65],"3,968":[66],"multi-speaker":[67,102],"audio":[68,122,191,211],"mixtures,":[69],"including":[70],"both":[71,101],"real-world":[72],"synthetic":[74],"scenarios,":[75],"paired":[76],"with":[77,120],"77k":[78],"multiple-choice":[79],"questions":[80],"probe":[82],"under":[84,173],"general":[85],"operating":[88],"modes.":[89],"In":[90],"addition,":[91],"we":[92,137],"propose":[93],"Selective":[94],"Efficacy":[95],"(SE),":[96],"novel":[98,145],"metric":[99],"capturing":[100],"comprehension":[103],"bystander-privacy":[105],"protection.":[106],"Our":[107],"evaluation":[108],"of":[109,130],"state-of-the-art":[110],"open-source":[111],"proprietary":[113],"LLMs":[114],"reveals":[115],"substantial":[116,164],"leakage,":[119],"strong":[121],"understanding":[123],"failing":[124],"translate":[126],"into":[127],"protection":[129],"privacy.":[132],"To":[133],"mitigate":[134],"this":[135],"gap,":[136],"also":[138],"present":[139],"Bystander":[140],"Privacy":[141],"Fine-Tuning":[142],"(BPFT),":[143],"training":[146],"pipeline":[147],"teaches":[149],"refuse":[152],"bystander-related":[153],"queries":[154],"without":[155,193],"degrading":[156],"main-speaker":[157],"comprehension.":[158],"show":[160],"BPFT":[162,198],"yields":[163],"gains,":[165],"achieving":[166],"absolute":[168,178],"47%":[169],"higher":[170,180],"accuracy":[172],"mode":[175],"16%":[179],"SE":[181],"compared":[182],"Gemini":[184],"2.5":[185],"Pro,":[186],"which":[187],"is":[188],"best":[190],"LLM":[192],"BPFT.":[194],"Together,":[195],"provide":[199],"systematic":[202],"framework":[203],"for":[204],"measuring":[205],"improving":[207],"LLMs.":[212]},"counts_by_year":[],"updated_date":"2026-10-29T09:56:32.521200","created_date":"2025-12-10T00:00:00"}