{"id":"https://openalex.org/W4412888284","doi":"https://doi.org/10.18653/v1/2025.findings-acl.534","title":"Unlocking Speech Instruction Data Potential with Query Rewriting","display_name":"Unlocking Speech Instruction Data Potential with Query Rewriting","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412888284","doi":"https://doi.org/10.18653/v1/2025.findings-acl.534"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.534","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.534","pdf_url":"https://aclanthology.org/2025.findings-acl.534.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.534.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5112264710","display_name":"Y.J. Hei","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yonghua Hei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5072900417","display_name":"Yibo Yan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yibo Yan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027064196","display_name":"Shuliang Liu","orcid":"https://orcid.org/0000-0002-9341-3116"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shuliang Liu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066119228","display_name":"Huiyu Zhou","orcid":"https://orcid.org/0000-0003-1634-9840"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huiyu Zhou","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5035516565","display_name":"Linfeng Zhang","orcid":"https://orcid.org/0000-0003-0452-4402"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Linfeng Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5043651092","display_name":"Xuming Hu","orcid":"https://orcid.org/0000-0001-6075-4224"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xuming Hu","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":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"10239","last_page":"10260"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.942300021648407,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.942300021648407,"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/T12031","display_name":"Speech and dialogue systems","score":0.9204000234603882,"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":0.9143999814987183,"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":0.8460448384284973},{"id":"https://openalex.org/keywords/rewriting","display_name":"Rewriting","score":0.8159180283546448},{"id":"https://openalex.org/keywords/query-language","display_name":"Query language","score":0.5334242582321167},{"id":"https://openalex.org/keywords/query-optimization","display_name":"Query optimization","score":0.47380632162094116},{"id":"https://openalex.org/keywords/rdf-query-language","display_name":"RDF query language","score":0.471293568611145},{"id":"https://openalex.org/keywords/query-by-example","display_name":"Query by Example","score":0.44977259635925293},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.3948614001274109},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.38280367851257324},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.3520638942718506},{"id":"https://openalex.org/keywords/web-search-query","display_name":"Web search query","score":0.3020530343055725},{"id":"https://openalex.org/keywords/web-query-classification","display_name":"Web query classification","score":0.2391863465309143},{"id":"https://openalex.org/keywords/search-engine","display_name":"Search engine","score":0.07637611031532288}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8460448384284973},{"id":"https://openalex.org/C154690210","wikidata":"https://www.wikidata.org/wiki/Q1668499","display_name":"Rewriting","level":2,"score":0.8159180283546448},{"id":"https://openalex.org/C192028432","wikidata":"https://www.wikidata.org/wiki/Q845739","display_name":"Query language","level":2,"score":0.5334242582321167},{"id":"https://openalex.org/C157692150","wikidata":"https://www.wikidata.org/wiki/Q2919848","display_name":"Query optimization","level":2,"score":0.47380632162094116},{"id":"https://openalex.org/C96956885","wikidata":"https://www.wikidata.org/wiki/Q6138701","display_name":"RDF query language","level":5,"score":0.471293568611145},{"id":"https://openalex.org/C194222762","wikidata":"https://www.wikidata.org/wiki/Q114486","display_name":"Query by Example","level":4,"score":0.44977259635925293},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.3948614001274109},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.38280367851257324},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3520638942718506},{"id":"https://openalex.org/C164120249","wikidata":"https://www.wikidata.org/wiki/Q995982","display_name":"Web search query","level":3,"score":0.3020530343055725},{"id":"https://openalex.org/C118689300","wikidata":"https://www.wikidata.org/wiki/Q7978614","display_name":"Web query classification","level":4,"score":0.2391863465309143},{"id":"https://openalex.org/C97854310","wikidata":"https://www.wikidata.org/wiki/Q19541","display_name":"Search engine","level":2,"score":0.07637611031532288}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.534","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.534","pdf_url":"https://aclanthology.org/2025.findings-acl.534.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.534","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.534","pdf_url":"https://aclanthology.org/2025.findings-acl.534.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":[{"score":0.6200000047683716,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320323537","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412888284.pdf","grobid_xml":"https://content.openalex.org/works/W4412888284.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W319014924","https://openalex.org/W2096359267","https://openalex.org/W2572349046","https://openalex.org/W2392799717","https://openalex.org/W3125756434","https://openalex.org/W2406556739","https://openalex.org/W2394844208","https://openalex.org/W4256058599","https://openalex.org/W2970853428","https://openalex.org/W2161787409"],"abstract_inverted_index":{"End-to-end":[0],"Large":[1,53],"Speech":[2],"Language":[3,54],"Models":[4,55],"(LSLMs)":[5],"demonstrate":[6],"strong":[7],"potential":[8],"in":[9,144,216],"response":[10],"latency":[11],"and":[12,40,76,92,112,165,223],"speech":[13,20,27,63,66,94,100,105,134,176,200],"comprehension":[14],"capabilities,":[15],"showcasing":[16],"general":[17],"intelligence":[18],"across":[19],"understanding":[21],"tasks.However,":[22],"the":[23,36,45,59,71,80,87,137,140,167],"ability":[24],"to":[25,35,57,64,70,102,127,133,136,163,173,210],"follow":[26],"instructions":[28,191],"has":[29,108],"not":[30],"been":[31],"fully":[32],"realized":[33],"due":[34,69,135],"lack":[37],"of":[38,62,90,139],"datasets":[39,96,107,178],"heavily":[41],"biased":[42],"training":[43,141],"tasks.Leveraging":[44],"rich":[46],"ASR":[47],"datasets,":[48],"previous":[49],"approaches":[50],"have":[51,119],"used":[52],"(LLMs)":[56],"continue":[58],"linguistic":[60],"information":[61],"construct":[65,103,174],"instruction":[67,95,106,132,177],"datasets.Yet,":[68],"gap":[72],"between":[73],"LLM-generated":[74],"results":[75],"real":[77],"human":[78,182],"responses,":[79],"continuation":[81],"methods":[82],"further":[83],"amplify":[84],"these":[85],"shortcomings.Given":[86],"high":[88],"costs":[89],"collecting":[91],"annotating":[93],"by":[97],"humans,":[98],"using":[99],"synthesis":[101,122,201],"largescale":[104],"become":[109],"a":[110,152],"balanced":[111],"robust":[113],"alternative.Although":[114],"modern":[115],"Text-To-Speech":[116],"(TTS)":[117],"models":[118,198],"achieved":[120],"near-human-level":[121],"quality,":[123],"it":[124,171],"is":[125],"challenging":[126],"appropriately":[128],"convert":[129],"out-ofdistribution":[130],"text":[131,190],"limitations":[138],"data":[142,206],"distribution":[143],"TTS":[145,197],"models.To":[146],"address":[147],"this":[148,186],"issue,":[149],"we":[150],"propose":[151],"query":[153],"rewriting":[154,217],"framework":[155],"with":[156],"multi-LLM":[157],"knowledge":[158,222],"fusion,":[159],"employing":[160],"multiple":[161],"agents":[162],"annotate":[164],"validate":[166],"synthesized":[168],"speech,":[169],"making":[170],"possible":[172],"high-quality":[175],"without":[179],"relying":[180],"on":[181],"annotation.Experiments":[183],"show":[184],"that":[185,219],"method":[187],"can":[188],"transform":[189],"into":[192],"distributions":[193],"more":[194],"suitable":[195],"for":[196,199],"through":[202],"zero-shot":[203],"rewriting,":[204],"increasing":[205],"usability":[207],"from":[208],"72%":[209],"93%.It":[211],"also":[212],"demonstrates":[213],"unique":[214],"advantages":[215],"tasks":[218],"require":[220],"complex":[221],"context-related":[224],"abilities.":[225]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}