{"id":"https://openalex.org/W7140121748","doi":"https://doi.org/10.18653/v1/2026.eacl-long.66","title":"Attacker\u2019s Noise Can Manipulate Your Audio-based LLM in the Real World","display_name":"Attacker\u2019s Noise Can Manipulate Your Audio-based LLM in the Real World","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7140121748","doi":"https://doi.org/10.18653/v1/2026.eacl-long.66"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.eacl-long.66","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.eacl-long.66","pdf_url":"https://aclanthology.org/2026.eacl-long.66.pdf","source":null,"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 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.eacl-long.66.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5048273558","display_name":"Vinu Sankar Sadasivan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Vinu Sankar Sadasivan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025450606","display_name":"Soheil Feizi","orcid":"https://orcid.org/0000-0003-0944-8242"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Soheil Feizi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075324202","display_name":"Rajiv Mathews","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rajiv Mathews","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130347612","display_name":"Lun Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lun Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"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":0.45227741,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1430","last_page":"1440"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11800","display_name":"User Authentication and Security Systems","score":0.29100000858306885,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T11800","display_name":"User Authentication and Security Systems","score":0.29100000858306885,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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":0.1120000034570694,"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":0.0835999995470047,"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/noise","display_name":"Noise (video)","score":0.5759999752044678},{"id":"https://openalex.org/keywords/background-noise","display_name":"Background noise","score":0.310699999332428},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.25519999861717224},{"id":"https://openalex.org/keywords/power","display_name":"Power (physics)","score":0.2502000033855438}],"concepts":[{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.5759999752044678},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5548999905586243},{"id":"https://openalex.org/C24890656","wikidata":"https://www.wikidata.org/wiki/Q82811","display_name":"Acoustics","level":1,"score":0.3314000070095062},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3257000148296356},{"id":"https://openalex.org/C100675267","wikidata":"https://www.wikidata.org/wiki/Q1371624","display_name":"Background noise","level":2,"score":0.310699999332428},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.28040000796318054},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.26249998807907104},{"id":"https://openalex.org/C24326235","wikidata":"https://www.wikidata.org/wiki/Q126095","display_name":"Electronic engineering","level":1,"score":0.25839999318122864},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.25519999861717224},{"id":"https://openalex.org/C163258240","wikidata":"https://www.wikidata.org/wiki/Q25342","display_name":"Power (physics)","level":2,"score":0.2502000033855438}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.eacl-long.66","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.eacl-long.66","pdf_url":"https://aclanthology.org/2026.eacl-long.66.pdf","source":null,"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 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.eacl-long.66","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.eacl-long.66","pdf_url":"https://aclanthology.org/2026.eacl-long.66.pdf","source":null,"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 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7140121748.pdf","grobid_xml":"https://content.openalex.org/works/W7140121748.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"This":[0],"paper":[1],"investigates":[2],"the":[3,62,67,73,92,96,99],"real-world":[4,79],"vulnerabilities":[5],"of":[6,75,98],"audio-based":[7],"large":[8],"language":[9],"models":[10],"(ALLMs),":[11],"such":[12,33],"as":[13,34],"Qwen2-Audio.We":[14],"first":[15],"demonstrate":[16],"that":[17,53],"an":[18],"adversary":[19],"can":[20,64],"craft":[21],"stealthy":[22],"audio":[23],"perturbations":[24],"to":[25,37,78],"manipulate":[26],"ALLMs":[27,63],"into":[28],"exhibiting":[29],"specific":[30],"targeted":[31],"behaviors,":[32],"eliciting":[35],"responses":[36],"wake-keywords":[38],"(e.g.,":[39,46],"\"Hey":[40],"Qwen\"),":[41],"or":[42],"triggering":[43],"harmful":[44],"behaviors":[45],"\"Change":[47],"my":[48],"calendar":[49],"event\").Subsequently,":[50],"we":[51,94],"show":[52],"playing":[54],"adversarial":[55,87],"background":[56],"noise":[57],"during":[58],"user":[59],"interaction":[60],"with":[61],"significantly":[65],"degrade":[66],"response":[68],"quality.Crucially,":[69],"our":[70],"research":[71],"illustrates":[72],"scalability":[74],"these":[76,86],"attacks":[77],"scenarios,":[80],"impacting":[81],"other":[82],"innocent":[83],"users":[84],"when":[85],"noises":[88],"are":[89],"played":[90],"through":[91],"air.Further,":[93],"discuss":[95],"transferability":[97],"attack":[100],"and":[101],"potential":[102],"defensive":[103],"measures.":[104]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-03-24T00:00:00"}
