{"id":"https://openalex.org/W4412887958","doi":"https://doi.org/10.18653/v1/2025.findings-acl.884","title":"Who Can Withstand Chat-Audio Attacks? An Evaluation Benchmark for Large Audio-Language Models","display_name":"Who Can Withstand Chat-Audio Attacks? An Evaluation Benchmark for Large Audio-Language Models","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412887958","doi":"https://doi.org/10.18653/v1/2025.findings-acl.884"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.884","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.884","pdf_url":"https://aclanthology.org/2025.findings-acl.884.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":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-acl.884.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5078658353","display_name":"Wanqi Yang","orcid":"https://orcid.org/0000-0001-6727-6077"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wanqi Yang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5010502049","display_name":"Yanda Li","orcid":"https://orcid.org/0009-0007-1123-9673"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yanda Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100869780","display_name":"Meng Fang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Meng Fang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087043856","display_name":"Yunchao Wei","orcid":"https://orcid.org/0000-0002-2812-8781"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yunchao Wei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5101656117","display_name":"Chen Ling","orcid":"https://orcid.org/0000-0002-3340-9342"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ling Chen","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":1.673,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.83323847,"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":"17205","last_page":"17220"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9398999810218811,"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":0.9398999810218811,"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/computer-science","display_name":"Computer science","score":0.7992445230484009},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.6493176221847534},{"id":"https://openalex.org/keywords/audio-over-ethernet","display_name":"Audio over Ethernet","score":0.5430097579956055},{"id":"https://openalex.org/keywords/audio-signal-processing","display_name":"Audio signal processing","score":0.4106362760066986},{"id":"https://openalex.org/keywords/multimedia","display_name":"Multimedia","score":0.3274853527545929},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.32153844833374023},{"id":"https://openalex.org/keywords/audio-signal","display_name":"Audio signal","score":0.273547500371933},{"id":"https://openalex.org/keywords/speech-coding","display_name":"Speech coding","score":0.2416868507862091}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7992445230484009},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.6493176221847534},{"id":"https://openalex.org/C36561122","wikidata":"https://www.wikidata.org/wiki/Q4819874","display_name":"Audio over Ethernet","level":5,"score":0.5430097579956055},{"id":"https://openalex.org/C127220857","wikidata":"https://www.wikidata.org/wiki/Q2719318","display_name":"Audio signal processing","level":4,"score":0.4106362760066986},{"id":"https://openalex.org/C49774154","wikidata":"https://www.wikidata.org/wiki/Q131765","display_name":"Multimedia","level":1,"score":0.3274853527545929},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.32153844833374023},{"id":"https://openalex.org/C64922751","wikidata":"https://www.wikidata.org/wiki/Q4650799","display_name":"Audio signal","level":3,"score":0.273547500371933},{"id":"https://openalex.org/C13895895","wikidata":"https://www.wikidata.org/wiki/Q3270773","display_name":"Speech coding","level":2,"score":0.2416868507862091},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0},{"id":"https://openalex.org/C13280743","wikidata":"https://www.wikidata.org/wiki/Q131089","display_name":"Geodesy","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.884","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.884","pdf_url":"https://aclanthology.org/2025.findings-acl.884.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":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-acl.884","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.884","pdf_url":"https://aclanthology.org/2025.findings-acl.884.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":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412887958.pdf","grobid_xml":"https://content.openalex.org/works/W4412887958.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2362963955","https://openalex.org/W4384080516","https://openalex.org/W195127310","https://openalex.org/W4254699594","https://openalex.org/W2141529408","https://openalex.org/W1512284021","https://openalex.org/W1526651343","https://openalex.org/W2109946194","https://openalex.org/W2115553177","https://openalex.org/W2350362445"],"abstract_inverted_index":{"Adversarial":[0],"audio":[1,37,54,66,139],"attacks":[2,67,140],"pose":[3],"a":[4,30],"significant":[5],"threat":[6],"to":[7,36,58,64,86],"the":[8,44,60,72,127,133,142,151,161],"growing":[9],"use":[10],"of":[11,53,62,74,135,138,144,154],"large":[12],"audiolanguage":[13],"models":[14],"(LALMs)":[15],"in":[16,68],"voice-based":[17],"human-machine":[18],"interactions.While":[19],"existing":[20],"research":[21],"focused":[22],"on":[23,126,141],"model-specific":[24],"adversarial":[25,38],"methods,":[26],"real-world":[27,96],"applications":[28],"demand":[29],"more":[31],"generalizable":[32],"and":[33,99,107,119],"universal":[34],"approach":[35],"attacks.In":[39],"this":[40],"paper,":[41],"we":[42,76],"introduce":[43],"Chat-Audio":[45],"Attacks":[46],"(CAA)":[47],"benchmark":[48],"including":[49,117],"four":[50,136],"distinct":[51,123],"types":[52,137],"attacks,":[55],"which":[56,94],"aims":[57],"explore":[59],"vulnerabilities":[61],"LALMs":[63,112],"these":[65,145],"conversational":[69,97],"scenarios.To":[70],"evaluate":[71,109],"robustness":[73],"LALMs,":[75],"propose":[77],"three":[78,122],"evaluation":[79,124],"strategies:":[80],"Standard":[81],"Evaluation,":[82,93,101],"utilizing":[83],"traditional":[84],"metrics":[85],"quantify":[87],"model":[88],"performance":[89,143],"under":[90],"attacks;":[91],"GPT-4o-Based":[92],"simulates":[95],"complexities;":[98],"Human":[100],"offering":[102],"insights":[103],"into":[104],"user":[105],"perception":[106],"trust.We":[108],"six":[110],"stateof-the-art":[111],"with":[113],"voice":[114],"interaction":[115],"capabilities,":[116],"Gemini-1.5-Pro,GPT-4o,":[118],"others,":[120],"using":[121],"methods":[125],"CAA":[128],"benchmark.Our":[129],"comprehensive":[130],"analysis":[131],"reveals":[132],"impact":[134],"models,":[146],"demonstrating":[147],"that":[148],"GPT-4o":[149],"exhibits":[150],"highest":[152],"level":[153],"resilience.Our":[155],"data":[156],"can":[157],"be":[158],"accessed":[159],"via":[160],"following":[162],"link:":[163],"CAA.":[164]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}