{"id":"https://openalex.org/W4416016960","doi":"https://doi.org/10.1145/3746252.3761269","title":"DistillCaps: Enhancing Audio-Language Alignment in Captioning via Retrieval-Augmented Knowledge Distillation","display_name":"DistillCaps: Enhancing Audio-Language Alignment in Captioning via Retrieval-Augmented Knowledge Distillation","publication_year":2025,"publication_date":"2025-11-07","ids":{"openalex":"https://openalex.org/W4416016960","doi":"https://doi.org/10.1145/3746252.3761269"},"language":null,"primary_location":{"id":"doi:10.1145/3746252.3761269","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3746252.3761269","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 34th ACM International Conference on Information and Knowledge Management","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Thinh Pham","orcid":"https://orcid.org/0009-0009-2465-6833"},"institutions":[{"id":"https://openalex.org/I123565023","display_name":"Vietnam National University Ho Chi Minh City","ror":"https://ror.org/00waaqh38","country_code":"VN","type":"education","lineage":["https://openalex.org/I123565023"]}],"countries":["VN"],"is_corresponding":false,"raw_author_name":"Thinh Pham","raw_affiliation_strings":["University of Science, Vietnam National University, Ho Chi Minh City, Vietnam"],"raw_orcid":"https://orcid.org/0009-0009-2465-6833","affiliations":[{"raw_affiliation_string":"University of Science, Vietnam National University, Ho Chi Minh City, Vietnam","institution_ids":["https://openalex.org/I123565023"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5058208255","display_name":"Tuong-Nghiem Diep","orcid":"https://orcid.org/0000-0001-7406-1250"},"institutions":[{"id":"https://openalex.org/I123565023","display_name":"Vietnam National University Ho Chi Minh City","ror":"https://ror.org/00waaqh38","country_code":"VN","type":"education","lineage":["https://openalex.org/I123565023"]}],"countries":["VN"],"is_corresponding":false,"raw_author_name":"Nghiem Diep","raw_affiliation_strings":["University of Science, Vietnam National University, Ho Chi Minh City, Vietnam"],"raw_orcid":"https://orcid.org/0000-0001-7406-1250","affiliations":[{"raw_affiliation_string":"University of Science, Vietnam National University, Ho Chi Minh City, Vietnam","institution_ids":["https://openalex.org/I123565023"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5081165986","display_name":"Lizi Liao","orcid":"https://orcid.org/0000-0002-9973-3305"},"institutions":[{"id":"https://openalex.org/I79891267","display_name":"Singapore Management University","ror":"https://ror.org/050qmg959","country_code":"SG","type":"education","lineage":["https://openalex.org/I79891267"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Lizi Liao","raw_affiliation_strings":["School of Computing and Information Systems, Singapore Management University, Singapore, Singapore"],"raw_orcid":"https://orcid.org/0000-0002-9973-3305","affiliations":[{"raw_affiliation_string":"School of Computing and Information Systems, Singapore Management University, Singapore, Singapore","institution_ids":["https://openalex.org/I79891267"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5051882105","display_name":"Binh T. Nguyen","orcid":"https://orcid.org/0000-0001-5249-9702"},"institutions":[{"id":"https://openalex.org/I123565023","display_name":"Vietnam National University Ho Chi Minh City","ror":"https://ror.org/00waaqh38","country_code":"VN","type":"education","lineage":["https://openalex.org/I123565023"]}],"countries":["VN"],"is_corresponding":false,"raw_author_name":"Binh T. Nguyen","raw_affiliation_strings":["AISIA Lab, Ho Chi Minh City, Vietnam and Univerisity of Science, Vietnam National University, Ho Chi Minh City, Vietnam"],"raw_orcid":"https://orcid.org/0000-0001-5249-9702","affiliations":[{"raw_affiliation_string":"AISIA Lab, Ho Chi Minh City, Vietnam and Univerisity of Science, Vietnam National University, Ho Chi Minh City, Vietnam","institution_ids":["https://openalex.org/I123565023"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.40477437,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"2346","last_page":"2356"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11309","display_name":"Music and Audio Processing","score":0.6589000225067139,"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/T11309","display_name":"Music and Audio Processing","score":0.6589000225067139,"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":0.1509000062942505,"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/T10860","display_name":"Speech and Audio Processing","score":0.08699999749660492,"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/closed-captioning","display_name":"Closed captioning","score":0.8783000111579895},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5867000222206116},{"id":"https://openalex.org/keywords/adapter","display_name":"Adapter (computing)","score":0.5807999968528748},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.5131000280380249},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.4726000130176544},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.45910000801086426},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.4332999885082245},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.41990000009536743}],"concepts":[{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.8783000111579895},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8264999985694885},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5867000222206116},{"id":"https://openalex.org/C177284502","wikidata":"https://www.wikidata.org/wiki/Q1005390","display_name":"Adapter (computing)","level":2,"score":0.5807999968528748},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.5131000280380249},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.484499990940094},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.4726000130176544},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.45910000801086426},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.45590001344680786},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4456000030040741},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.4332999885082245},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.41990000009536743},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.40709999203681946},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3815999925136566},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.38019999861717224},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.3707999885082245},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.334199994802475},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.32679998874664307},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.3034999966621399},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.27720001339912415},{"id":"https://openalex.org/C75165309","wikidata":"https://www.wikidata.org/wiki/Q2258979","display_name":"Search engine indexing","level":2,"score":0.2540999948978424},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.25119999051094055}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3746252.3761269","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3746252.3761269","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 34th ACM International Conference on Information and Knowledge Management","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":28,"referenced_works":["https://openalex.org/W1956340063","https://openalex.org/W1972475755","https://openalex.org/W1978968298","https://openalex.org/W2097999418","https://openalex.org/W2101132289","https://openalex.org/W2506483933","https://openalex.org/W2526050071","https://openalex.org/W2593116425","https://openalex.org/W2916103538","https://openalex.org/W2945761034","https://openalex.org/W2949376505","https://openalex.org/W2964213897","https://openalex.org/W2969472291","https://openalex.org/W2982669287","https://openalex.org/W3015591594","https://openalex.org/W3094550259","https://openalex.org/W3160577380","https://openalex.org/W3187963534","https://openalex.org/W4226442948","https://openalex.org/W4286278015","https://openalex.org/W4372266552","https://openalex.org/W4372340819","https://openalex.org/W4385823080","https://openalex.org/W4387609319","https://openalex.org/W4392902953","https://openalex.org/W4392903210","https://openalex.org/W4392904001","https://openalex.org/W4408352941"],"related_works":[],"abstract_inverted_index":{"Automated":[0],"audio":[1,77,104,117,127],"captioning":[2,141,182],"(AAC)":[3],"benefits":[4,147],"from":[5,103],"incorporating":[6],"external":[7],"context":[8],"to":[9,26,45,84,94,99,119,199],"interpret":[10],"complex":[11],"sounds,":[12],"but":[13],"doing":[14],"so":[15],"with":[16,163],"retrieval-augmented":[17],"generation":[18],"(RAG)":[19],"at":[20,173],"inference":[21],"is":[22,92,138],"sometimes":[23],"infeasible":[24],"due":[25],"data":[27],"availability":[28],"or":[29,164],"incurs":[30],"significant":[31],"latency":[32],"and":[33,79,157,187,219],"complexity.":[34],"We":[35,106],"propose":[36],"DistillCaps,":[37],"a":[38,64,89,109],"novel":[39],"training-time":[40],"framework":[41],"that":[42,143],"leverages":[43],"RAG":[44],"guide":[46],"knowledge":[47],"distillation":[48,214],"for":[49,75,82,215],"improved":[50],"audio-language":[51,216],"alignment,":[52],"while":[53],"lessening":[54],"the":[55,116,124,133,202,206,210],"reliance":[56],"on":[57,201,205],"retrieval":[58,172],"during":[59],"inference.":[60],"In":[61],"our":[62,177],"framework,":[63],"RAG-equipped":[65],"teacher":[66],"model":[67,91,142,179],"retrieves":[68],"relevant":[69],"textual":[70],"information":[71],"(e.g.,":[72],"similar":[73],"captions)":[74],"each":[76],"clip":[78],"uses":[80],"it":[81,192],"training":[83],"generate":[85],"context-enriched":[86],"captions.":[87],"Simultaneously,":[88],"student":[90],"trained":[93],"imitate":[95],"this":[96],"teacher,":[97],"learning":[98],"produce":[100],"high-quality":[101],"captions":[102],"alone.":[105],"further":[107],"introduce":[108],"Fast":[110],"Fourier":[111],"Transform":[112],"(FFT)":[113],"adapter":[114],"in":[115],"encoder":[118],"inject":[120],"frequency-domain":[121],"features,":[122],"enhancing":[123],"quality":[125],"of":[126,212],"representations":[128],"before":[129],"feeding":[130],"them":[131],"into":[132],"language":[134],"model.":[135],"The":[136],"result":[137],"an":[139],"efficient":[140],"retains":[144],"RAG's":[145],"contextual":[146],"without":[148],"its":[149],"deployment":[150],"overhead.":[151],"On":[152],"standard":[153],"AAC":[154],"benchmarks":[155],"(AudioCaps":[156],"Clotho),":[158],"DistillCaps":[159],"achieves":[160],"performance":[161],"competitive":[162],"exceeding":[165],"prior":[166],"RAG-based":[167],"systems":[168],"despite":[169],"using":[170],"no":[171],"test":[174],"time.":[175],"Notably,":[176],"distilled":[178],"matches":[180],"state-of-the-art":[181],"results":[183],"under":[184],"real-time":[185],"settings,":[186],"when":[188],"optionally":[189],"allowing":[190],"retrieval,":[191],"even":[193],"outperforms":[194],"previous":[195],"models":[196],"by":[197],"up":[198],"4%":[200],"Clotho":[203],"benchmark":[204],"in-distribution":[207],"setting,":[208],"demonstrating":[209],"effectiveness":[211],"RAG-guided":[213],"alignment.":[217],"Code":[218],"dataset":[220],"are":[221],"available":[222],"here.":[223],"https://github.com/pgthinh/DistillCaps.":[224]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-11-08T00:00:00"}