{"id":"https://openalex.org/W7127707891","doi":"https://doi.org/10.48550/arxiv.2602.02557","title":"The Alignment Curse: Modality Alignment Supercharges Audio Attacks via Text Transfer","display_name":"The Alignment Curse: Modality Alignment Supercharges Audio Attacks via Text Transfer","publication_year":2026,"publication_date":"2026-01-30","ids":{"openalex":"https://openalex.org/W7127707891","doi":"https://doi.org/10.48550/arxiv.2602.02557"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2602.02557","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.02557","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,"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2602.02557","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5124931386","display_name":"Yupeng Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Yupeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060684352","display_name":"Junchi Yu","orcid":"https://orcid.org/0000-0003-4118-3248"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Junchi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125081008","display_name":"Aoxi Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Aoxi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5146919776","display_name":"Baoyuan Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Baoyuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125066685","display_name":"Philip Torr","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Torr, Philip","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5091670711","display_name":"Adel Bibi","orcid":"https://orcid.org/0000-0002-6169-3918"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bibi, Adel","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":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.45080000162124634,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.45080000162124634,"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/T10028","display_name":"Topic Modeling","score":0.09759999811649323,"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/T11241","display_name":"Advanced Malware Detection Techniques","score":0.06909999996423721,"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/transferability","display_name":"Transferability","score":0.5863999724388123},{"id":"https://openalex.org/keywords/modality","display_name":"Modality (human\u2013computer interaction)","score":0.5684999823570251},{"id":"https://openalex.org/keywords/similarity","display_name":"Similarity (geometry)","score":0.545799970626831},{"id":"https://openalex.org/keywords/transfer","display_name":"Transfer (computing)","score":0.4772000014781952},{"id":"https://openalex.org/keywords/bridge","display_name":"Bridge (graph theory)","score":0.4415000081062317},{"id":"https://openalex.org/keywords/modalities","display_name":"Modalities","score":0.4375},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.35760000348091125},{"id":"https://openalex.org/keywords/term","display_name":"Term (time)","score":0.34630000591278076}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7904000282287598},{"id":"https://openalex.org/C61272859","wikidata":"https://www.wikidata.org/wiki/Q7834031","display_name":"Transferability","level":3,"score":0.5863999724388123},{"id":"https://openalex.org/C2780226545","wikidata":"https://www.wikidata.org/wiki/Q6888030","display_name":"Modality (human\u2013computer interaction)","level":2,"score":0.5684999823570251},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.545799970626831},{"id":"https://openalex.org/C2776175482","wikidata":"https://www.wikidata.org/wiki/Q1195816","display_name":"Transfer (computing)","level":2,"score":0.4772000014781952},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4584999978542328},{"id":"https://openalex.org/C100776233","wikidata":"https://www.wikidata.org/wiki/Q2532492","display_name":"Bridge (graph theory)","level":2,"score":0.4415000081062317},{"id":"https://openalex.org/C2779903281","wikidata":"https://www.wikidata.org/wiki/Q6888026","display_name":"Modalities","level":2,"score":0.4375},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.40639999508857727},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3962000012397766},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.35760000348091125},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.34630000591278076},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.31029999256134033},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.3089999854564667},{"id":"https://openalex.org/C13355873","wikidata":"https://www.wikidata.org/wiki/Q2920850","display_name":"Connection (principal bundle)","level":2,"score":0.2863999903202057},{"id":"https://openalex.org/C8854915","wikidata":"https://www.wikidata.org/wiki/Q4350200","display_name":"Information transfer","level":2,"score":0.2761000096797943},{"id":"https://openalex.org/C2780910867","wikidata":"https://www.wikidata.org/wiki/Q1952416","display_name":"Multimodality","level":2,"score":0.275299996137619},{"id":"https://openalex.org/C2778012447","wikidata":"https://www.wikidata.org/wiki/Q1034415","display_name":"Scope (computer science)","level":2,"score":0.26649999618530273},{"id":"https://openalex.org/C2779662365","wikidata":"https://www.wikidata.org/wiki/Q5416694","display_name":"Event (particle physics)","level":2,"score":0.2655999958515167},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.2603999972343445},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.25870001316070557},{"id":"https://openalex.org/C59656382","wikidata":"https://www.wikidata.org/wiki/Q191536","display_name":"Conjunction (astronomy)","level":2,"score":0.2540000081062317},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.2531000077724457},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.25290000438690186}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2602.02557","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.02557","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,"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2602.02557","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.02557","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,"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"display_name":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16","score":0.815768837928772}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Recent":[0],"advances":[1],"in":[2,159],"end-to-end":[3],"trained":[4],"omni-models":[5,116],"have":[6],"substantially":[7],"improved":[8],"audio":[9,51,123,126,132,161],"capabilities":[10],"by":[11,101],"strengthening":[12],"text-audio":[13],"modality":[14,80,171,189],"alignment.":[15],"However,":[16],"whether":[17],"such":[18],"alignment":[19,81,172,190],"inadvertently":[20],"facilitates":[21],"the":[22,59,67,168,185],"transfer":[23,48,85,174],"of":[24,86,110],"safety":[25,52,162],"vulnerabilities":[26,154],"across":[27,176],"modalities":[28],"remains":[29],"underexplored.":[30],"This":[31,150],"question":[32],"is":[33],"critical":[34],"as":[35],"text-based":[36,153],"jailbreak":[37],"attacks":[38,87,133],"are":[39],"considerably":[40],"more":[41,83,193],"mature":[42],"than":[43],"audio-based":[44,141],"ones;":[45],"if":[46],"they":[47],"systematically,":[49],"current":[50],"evaluations":[53],"may":[54],"underestimate":[55],"risks":[56],"originating":[57],"from":[58,88],"text":[60,89,120],"modality.":[61],"In":[62],"this":[63,102],"paper,":[64],"we":[65,104,165],"introduce":[66],"Alignment":[68,186],"Curse,":[69],"a":[70,93,106,144,156],"formally":[71],"characterized":[72],"and":[73,98,125,137,173,179],"empirically":[74,166],"validated":[75],"principle":[76],"showing":[77],"that":[78,130,152],"stronger":[79],"enables":[82],"effective":[84,194],"to":[90,192],"audio,":[91],"revealing":[92],"fundamental":[94],"tension":[95],"between":[96,170],"capability":[97],"safety.":[99],"Motivated":[100],"principle,":[103],"conduct":[105],"comprehensive":[107],"black-box":[108],"evaluation":[109],"three":[111],"attack":[112,177,196],"categories":[113],"on":[114],"recent":[115],"(e.g.,":[117],"Qwen2.5-Omni,":[118],"Qwen3-Omni):":[119],"attacks,":[121,124,142],"text-transferred":[122,131],"attacks.":[127],"We":[128],"find":[129],"perform":[134],"comparably":[135],"to,":[136],"often":[138],"better":[139],"than,":[140],"exhibiting":[143],"clear":[145],"advantage":[146],"under":[147],"audio-only":[148],"access.":[149],"suggests":[151],"play":[155],"pivotal":[157],"role":[158],"shaping":[160],"risks.":[163],"Finally,":[164],"analyze":[167],"relationship":[169],"effectiveness":[175],"methods":[178],"models,":[180],"observing":[181],"consistent":[182],"support":[183],"for":[184],"Curse:":[187],"tighter":[188],"leads":[191],"cross-modality":[195],"transfer.":[197]},"counts_by_year":[],"updated_date":"2026-08-16T07:02:28.622633","created_date":"2026-02-06T00:00:00"}