{"id":"https://openalex.org/W7204896830","doi":"https://doi.org/10.48550/arxiv.2608.30329","title":"Ouroboros: Self-Referential Backdoor Attacks on Speech Enhancement via Clean Audio Triggers","display_name":"Ouroboros: Self-Referential Backdoor Attacks on Speech Enhancement via Clean Audio Triggers","publication_year":2026,"publication_date":"2026-08-31","ids":{"openalex":"https://openalex.org/W7204896830","doi":"https://doi.org/10.48550/arxiv.2608.30329"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.30329","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.30329","pdf_url":"https://arxiv.org/pdf/2608.30329","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":null,"license_id":null,"version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2608.30329","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5148844900","display_name":"Yunjie Zhou","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Yunjie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5148835677","display_name":"Yuheng Huang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Yuheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5016795475","display_name":"Diqun Yan","orcid":"https://orcid.org/0000-0002-5241-7276","observed_orcids":["https://orcid.org/0000-0002-5241-7276"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yan, Diqun","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":0.9990000128746033,"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":0.9990000128746033,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.00019999999494757503,"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/T11424","display_name":"Security and Verification in Computing","score":9.999999747378752e-05,"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/backdoor-attacks","display_name":"backdoor attacks","score":0.9750000238418579},{"id":"https://openalex.org/keywords/speech-enhancement","display_name":"speech enhancement","score":0.9229999780654907},{"id":"https://openalex.org/keywords/adversarial-robustness","display_name":"adversarial robustness","score":0.6510000228881836},{"id":"https://openalex.org/keywords/defense-evasion","display_name":"defense evasion","score":0.531000018119812},{"id":"https://openalex.org/keywords/physical-world-attacks","display_name":"physical-world attacks","score":0.5619999766349792}],"concepts":[{"id":"https://openalex.org/C2781045450","wikidata":"https://www.wikidata.org/wiki/Q254569","display_name":"Backdoor","level":2,"score":0.9847999811172485},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7179999947547913},{"id":"https://openalex.org/C2776182073","wikidata":"https://www.wikidata.org/wiki/Q7575395","display_name":"Speech enhancement","level":3,"score":0.5784000158309937},{"id":"https://openalex.org/C95713431","wikidata":"https://www.wikidata.org/wiki/Q631425","display_name":"Vulnerability (computing)","level":2,"score":0.4828000068664551},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4668000042438507},{"id":"https://openalex.org/C2779679103","wikidata":"https://www.wikidata.org/wiki/Q5251805","display_name":"Degradation (telecommunications)","level":2,"score":0.3273000121116638},{"id":"https://openalex.org/C204201278","wikidata":"https://www.wikidata.org/wiki/Q1332614","display_name":"Voice activity detection","level":3,"score":0.3089999854564667},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.2919999957084656},{"id":"https://openalex.org/C13895895","wikidata":"https://www.wikidata.org/wiki/Q3270773","display_name":"Speech coding","level":2,"score":0.2897999882698059},{"id":"https://openalex.org/C2776608160","wikidata":"https://www.wikidata.org/wiki/Q4785462","display_name":"Natural (archaeology)","level":2,"score":0.27309998869895935},{"id":"https://openalex.org/C2776639384","wikidata":"https://www.wikidata.org/wiki/Q840396","display_name":"Ideal (ethics)","level":2,"score":0.27250000834465027},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.267300009727478}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.30329","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.30329","pdf_url":"https://arxiv.org/pdf/2608.30329","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":null,"license_id":null,"version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2608.30329","doi":"https://doi.org/10.48550/arxiv.2608.30329","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.30329","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2608.30329","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.30329","pdf_url":"https://arxiv.org/pdf/2608.30329","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":null,"license_id":null,"version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4304091911","https://openalex.org/W7139926047","https://openalex.org/W4408538350","https://openalex.org/W4402352499","https://openalex.org/W4315705628","https://openalex.org/W7125975307","https://openalex.org/W4398757423","https://openalex.org/W7165665597","https://openalex.org/W7223083470","https://openalex.org/W4409374173"],"abstract_inverted_index":{"Speech":[0],"enhancement":[1,45,66],"models":[2,67,94],"are":[3,24],"widely":[4],"deployed":[5],"as":[6,68],"frontend":[7],"modules":[8],"in":[9],"real-time":[10],"speech":[11,44,65],"services,":[12],"yet":[13],"their":[14],"vulnerability":[15],"to":[16,26,114],"backdoor":[17,22,55],"attacks":[18,117],"remains":[19,119],"unexplored.":[20],"Existing":[21],"methods":[23],"confined":[25],"classification":[27],"tasks":[28],"and":[29,95,118,124],"rely":[30],"on":[31,92],"active":[32],"trigger":[33,77],"injection,":[34],"an":[35],"assumption":[36],"incompatible":[37],"with":[38,88],"the":[39,60,109],"passive":[40],"processing":[41],"nature":[42],"of":[43,64],"models.":[46],"In":[47],"this":[48],"paper,":[49],"we":[50],"propose":[51],"Ouroboros,":[52],"a":[53],"novel":[54],"attack":[56,85],"framework":[57],"that":[58,100],"leverages":[59],"ideal":[61],"clean":[62,104],"outputs":[63],"natural":[69],"triggers,":[70],"enabling":[71],"inference-time":[72],"activation":[73],"without":[74],"any":[75],"external":[76],"injection.":[78],"Extensive":[79],"evaluations":[80],"show":[81],"Ouroboros":[82,112],"achieves":[83],"near-perfect":[84],"success":[86],"rates":[87],"minimal":[89],"performance":[90],"degradation":[91],"diverse":[93],"datasets.":[96],"Physical-world":[97],"validations":[98],"confirm":[99],"naturally":[101],"recorded,":[102],"unaltered":[103],"audio":[105],"can":[106],"reliably":[107],"activate":[108],"backdoor.":[110],"Moreover,":[111],"generalizes":[113],"targeted":[115],"content-tampering":[116],"effective":[120],"against":[121],"common":[122],"filtering":[123],"finetuning":[125],"defenses.":[126]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-09-02T00:00:00"}