{"id":"https://openalex.org/W4403913090","doi":"https://doi.org/10.1145/3678957.3685705","title":"DoubleDistillation: Enhancing LLMs for Informal Text Analysis using Multistage Knowledge Distillation from Speech and Text","display_name":"DoubleDistillation: Enhancing LLMs for Informal Text Analysis using Multistage Knowledge Distillation from Speech and Text","publication_year":2024,"publication_date":"2024-10-30","ids":{"openalex":"https://openalex.org/W4403913090","doi":"https://doi.org/10.1145/3678957.3685705"},"language":"en","primary_location":{"id":"doi:10.1145/3678957.3685705","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3678957.3685705","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"International Conference on Multimodel Interaction","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":"https://openalex.org/A5017111048","display_name":"Fatema Hasan","orcid":"https://orcid.org/0000-0001-9722-4243"},"institutions":[{"id":"https://openalex.org/I79272384","display_name":"University of Maryland, Baltimore County","ror":"https://ror.org/02qskvh78","country_code":"US","type":"education","lineage":["https://openalex.org/I79272384"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Fatema Hasan","raw_affiliation_strings":["Department of Information Systems, University of Maryland, Baltimore County, United States"],"raw_orcid":"https://orcid.org/0000-0001-9722-4243","affiliations":[{"raw_affiliation_string":"Department of Information Systems, University of Maryland, Baltimore County, United States","institution_ids":["https://openalex.org/I79272384"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5035840897","display_name":"Yulong Li","orcid":"https://orcid.org/0000-0003-3412-0732"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yulong Li","raw_affiliation_strings":["IBM Research AI, United States"],"raw_orcid":"https://orcid.org/0000-0003-3412-0732","affiliations":[{"raw_affiliation_string":"IBM Research AI, United States","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5010326834","display_name":"James R. Foulds","orcid":"https://orcid.org/0000-0003-0935-4182"},"institutions":[{"id":"https://openalex.org/I79272384","display_name":"University of Maryland, Baltimore County","ror":"https://ror.org/02qskvh78","country_code":"US","type":"education","lineage":["https://openalex.org/I79272384"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"James R. Foulds","raw_affiliation_strings":["Department of Information Systems, University of Maryland, Baltimore County, United States"],"raw_orcid":"https://orcid.org/0000-0003-0935-4182","affiliations":[{"raw_affiliation_string":"Department of Information Systems, University of Maryland, Baltimore County, United States","institution_ids":["https://openalex.org/I79272384"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048111120","display_name":"Shimei Pan","orcid":"https://orcid.org/0000-0002-5989-8543"},"institutions":[{"id":"https://openalex.org/I79272384","display_name":"University of Maryland, Baltimore County","ror":"https://ror.org/02qskvh78","country_code":"US","type":"education","lineage":["https://openalex.org/I79272384"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shimei Pan","raw_affiliation_strings":["Department of Information Systems, University of Maryland, Baltimore County, United States"],"raw_orcid":"https://orcid.org/0000-0002-5989-8543","affiliations":[{"raw_affiliation_string":"Department of Information Systems, University of Maryland, Baltimore County, United States","institution_ids":["https://openalex.org/I79272384"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5069125151","display_name":"Bishwaranjan Bhattacharjee","orcid":"https://orcid.org/0009-0009-7097-4891"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bishwaranjan Bhattacharjee","raw_affiliation_strings":["IBM Research AI, United States"],"raw_orcid":"https://orcid.org/0009-0009-7097-4891","affiliations":[{"raw_affiliation_string":"IBM Research AI, United States","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"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":"526","last_page":"535"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","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/T10028","display_name":"Topic Modeling","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":0.9945999979972839,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9896000027656555,"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/distillation","display_name":"Distillation","score":0.6759477853775024},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6141847372055054},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.446435809135437},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3521985411643982},{"id":"https://openalex.org/keywords/chemistry","display_name":"Chemistry","score":0.11863502860069275},{"id":"https://openalex.org/keywords/chromatography","display_name":"Chromatography","score":0.09055498242378235}],"concepts":[{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.6759477853775024},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6141847372055054},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.446435809135437},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3521985411643982},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.11863502860069275},{"id":"https://openalex.org/C43617362","wikidata":"https://www.wikidata.org/wiki/Q170050","display_name":"Chromatography","level":1,"score":0.09055498242378235}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3678957.3685705","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3678957.3685705","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"International Conference on Multimodel Interaction","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.4399999976158142}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":19,"referenced_works":["https://openalex.org/W1861492603","https://openalex.org/W2062249044","https://openalex.org/W2123442489","https://openalex.org/W2146574880","https://openalex.org/W2212660284","https://openalex.org/W2805744755","https://openalex.org/W2883409523","https://openalex.org/W2963855133","https://openalex.org/W2964051877","https://openalex.org/W2997006708","https://openalex.org/W3088631780","https://openalex.org/W3177777394","https://openalex.org/W3203532272","https://openalex.org/W3209059054","https://openalex.org/W3209984917","https://openalex.org/W3212867526","https://openalex.org/W4224903019","https://openalex.org/W4281701185","https://openalex.org/W6601141708"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W3204019825"],"abstract_inverted_index":{"Traditional":[0],"large":[1,186],"language":[2,48,78,99,154],"models":[3,28,34,54,79],"(LLMs)":[4],"leverage":[5],"extensive":[6],"text":[7,58,92,160,203],"corpora":[8,93],"but":[9],"lack":[10],"access":[11],"to":[12,80,97,126,151],"acoustic":[13,84,170],"and":[14,59,85,103,171,179],"para-linguistic":[15],"cues":[16,173],"present":[17],"in":[18,25,95,138,174,184,200],"speech.":[19],"There":[20],"is":[21,42,149],"a":[22,69,118,128,152,158,185],"growing":[23],"interest":[24],"enhancing":[26],"text-based":[27],"with":[29],"audio":[30,60,111],"information.":[31],"However,":[32],"current":[33],"often":[35,55],"require":[36,56],"an":[37,110,134],"aligned":[38,135,176],"audio-text":[39,136,177],"dataset":[40,137],"which":[41],"frequently":[43],"much":[44],"smaller":[45],"than":[46],"typical":[47,98],"model":[49,100,123,132,155],"training":[50,101],"corpora.":[51],"Moreover,":[52],"these":[53],"both":[57,168],"streams":[61],"during":[62,113],"inference/testing.":[63,114],"In":[64,142],"this":[65,193],"study,":[66],"we":[67,116],"introduce":[68],"novel":[70],"two-stage":[71,164],"knowledge":[72,148,183],"distillation":[73],"(KD)":[74],"approach":[75],"that":[76],"enables":[77],"(a)":[81],"incorporate":[82],"rich":[83],"paralinguistic":[86,172],"information":[87],"from":[88],"speech,":[89],"(b)":[90],"utilize":[91],"comparable":[94],"size":[96],"data,":[102],"(c)":[104],"support":[105],"text-only":[106,187],"analysis":[107],"without":[108],"requiring":[109],"stream":[112],"Specifically,":[115],"employ":[117],"pre-trained":[119],"speech":[120],"embedding":[121],"teacher":[122],"(OpenAI":[124],"Whisper)":[125],"train":[127],"Teacher":[129],"Assistant":[130],"(TA)":[131],"on":[133,157,190],"the":[139,143,146,169,175,180],"first":[140],"stage.":[141],"second":[144],"stage,":[145],"TA\u2019s":[147],"transferred":[150],"student":[153],"trained":[156],"conventional":[159],"dataset.":[161,188],"Thus,":[162],"our":[163,191],"KD":[165],"method":[166],"leverages":[167],"data":[178],"nuanced":[181],"linguistic":[182],"Based":[189],"evaluation,":[192],"DoubleDistillation":[194],"system":[195],"consistently":[196],"outperforms":[197],"traditional":[198],"LLMs":[199],"15":[201],"informal":[202],"understanding":[204],"tasks.":[205]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}