{"id":"https://openalex.org/W7166847393","doi":"https://doi.org/10.18653/v1/2026.findings-acl.967","title":"Beyond Transcription: Unified Audio Schema for Perception-Aware AudioLLMs","display_name":"Beyond Transcription: Unified Audio Schema for Perception-Aware AudioLLMs","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166847393","doi":"https://doi.org/10.18653/v1/2026.findings-acl.967"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.967","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.967","pdf_url":"https://aclanthology.org/2026.findings-acl.967.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 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.findings-acl.967.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5133704162","display_name":"Linhao Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Linhao Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102021469","display_name":"Yuhan Song","orcid":"https://orcid.org/0009-0002-6911-7955"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuhan Song","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139823058","display_name":"Aiwei Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Aiwei Liu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139719358","display_name":"Chuhan Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chuhan Wu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048880167","display_name":"Sijun Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sijun Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139802621","display_name":"Wei Jia","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wei Jia","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139845221","display_name":"Yuan Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuan Liu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139832433","display_name":"Houfeng Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Houfeng Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5084534688","display_name":"Zhou Xiao","orcid":"https://orcid.org/0000-0001-5263-365X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou Xiao","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":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.86341776,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"19375","last_page":"19390"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.541100025177002,"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"}},"topics":[{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.541100025177002,"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/T11309","display_name":"Music and Audio Processing","score":0.16750000417232513,"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":0.08110000193119049,"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/schema","display_name":"Schema (genetic algorithms)","score":0.3862999975681305},{"id":"https://openalex.org/keywords/sketch","display_name":"Sketch","score":0.265500009059906},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.2596000134944916},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.241799995303154}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6868000030517578},{"id":"https://openalex.org/C52146309","wikidata":"https://www.wikidata.org/wiki/Q7431116","display_name":"Schema (genetic algorithms)","level":2,"score":0.3862999975681305},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3456999957561493},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3434000015258789},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.26840001344680786},{"id":"https://openalex.org/C2779231336","wikidata":"https://www.wikidata.org/wiki/Q7534724","display_name":"Sketch","level":2,"score":0.265500009059906},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.2596000134944916},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.241799995303154},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.23729999363422394},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.23350000381469727}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.967","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.967","pdf_url":"https://aclanthology.org/2026.findings-acl.967.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 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.967","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.967","pdf_url":"https://aclanthology.org/2026.findings-acl.967.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 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166847393.pdf","grobid_xml":"https://content.openalex.org/works/W7166847393.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Recent":[0],"Audio":[1,57],"Large":[2],"Language":[3],"Models":[4],"(Au-dioLLMs)":[5],"exhibit":[6],"a":[7,28,60,78],"striking":[8],"performance":[9],"inversion:":[10],"while":[11,137],"excelling":[12],"at":[13,148],"complex":[14],"reasoning":[15,140],"tasks,":[16],"they":[17],"consistently":[18],"underperform":[19],"on":[20,114,130],"fine-grained":[21,126],"acoustic":[22,48,85],"perception.We":[23],"attribute":[24],"this":[25,100],"gap":[26],"to":[27,43,106],"fundamental":[29],"limitation":[30],"of":[31,99],"ASRcentric":[32],"training,":[33],"which":[34],"provides":[35],"precise":[36],"linguistic":[37],"targets":[38],"but":[39],"implicitly":[40],"teaches":[41],"models":[42,136],"suppress":[44],"paralinguistic":[45],"cues":[46],"and":[47,62,75,109,117,143],"events":[49],"as":[50],"noise.To":[51],"address":[52],"this,":[53],"we":[54],"propose":[55],"Unified":[56],"Schema":[58],"(UAS),":[59],"holistic":[61],"structured":[63],"supervision":[64,101],"framework":[65],"that":[66,93,120],"organizes":[67],"audio":[68],"information":[69],"into":[70],"three":[71],"explicit":[72],"components-Transcription,":[73],"Paralinguistics,":[74],"Non-linguistic":[76],"Events-within":[77],"unified":[79],"JSON":[80],"format.This":[81],"design":[82],"achieves":[83],"comprehensive":[84],"coverage":[86],"without":[87],"sacrificing":[88],"the":[89,97,133],"tight":[90],"audio-text":[91],"alignment":[92],"enables":[94],"reasoning.We":[95],"validate":[96],"effectiveness":[98],"strategy":[102],"by":[103,128],"applying":[104],"it":[105],"both":[107],"discrete":[108],"continuous":[110],"AudioLLM":[111],"architectures.Extensive":[112],"experiments":[113],"MMSU,":[115],"MMAR,":[116],"MMAU":[118],"demonstrate":[119],"UAS-Audio":[121],"yields":[122],"consistent":[123],"improvements,":[124],"boosting":[125],"perception":[127],"10.9%":[129],"MMSU":[131],"over":[132],"same-size":[134],"state-of-the-art":[135],"preserving":[138],"robust":[139],"capabilities.Our":[141],"code":[142],"model":[144],"are":[145],"publicly":[146],"available":[147],"https://github.com/":[149],"Tencent/Unified_Audio_Schema.":[150]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}