{"id":"https://openalex.org/W4401043057","doi":"https://doi.org/10.18653/v1/2024.findings-naacl.199","title":"DiLM: Distilling Dataset into Language Model for Text-level Dataset Distillation","display_name":"DiLM: Distilling Dataset into Language Model for Text-level Dataset Distillation","publication_year":2024,"publication_date":"2024-01-01","ids":{"openalex":"https://openalex.org/W4401043057","doi":"https://doi.org/10.18653/v1/2024.findings-naacl.199"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2024.findings-naacl.199","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2024.findings-naacl.199","pdf_url":"https://aclanthology.org/2024.findings-naacl.199.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: NAACL 2024","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2024.findings-naacl.199.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5111702849","display_name":"Aru Maekawa","orcid":null},"institutions":[{"id":"https://openalex.org/I114531698","display_name":"Tokyo Institute of Technology","ror":"https://ror.org/0112mx960","country_code":"JP","type":"education","lineage":["https://openalex.org/I114531698"]}],"countries":["JP"],"is_corresponding":true,"raw_author_name":"Aru Maekawa","raw_affiliation_strings":["Tokyo Institute of Technology"],"affiliations":[{"raw_affiliation_string":"Tokyo Institute of Technology","institution_ids":["https://openalex.org/I114531698"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5010134544","display_name":"Satoshi Kosugi","orcid":"https://orcid.org/0000-0001-7556-9072"},"institutions":[{"id":"https://openalex.org/I114531698","display_name":"Tokyo Institute of Technology","ror":"https://ror.org/0112mx960","country_code":"JP","type":"education","lineage":["https://openalex.org/I114531698"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Satoshi Kosugi","raw_affiliation_strings":["Tokyo Institute of Technology"],"affiliations":[{"raw_affiliation_string":"Tokyo Institute of Technology","institution_ids":["https://openalex.org/I114531698"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5069989297","display_name":"Kotaro Funakoshi","orcid":"https://orcid.org/0000-0002-4529-4634"},"institutions":[{"id":"https://openalex.org/I114531698","display_name":"Tokyo Institute of Technology","ror":"https://ror.org/0112mx960","country_code":"JP","type":"education","lineage":["https://openalex.org/I114531698"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Kotaro Funakoshi","raw_affiliation_strings":["Tokyo Institute of Technology"],"affiliations":[{"raw_affiliation_string":"Tokyo Institute of Technology","institution_ids":["https://openalex.org/I114531698"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5035876897","display_name":"Manabu Okumura","orcid":null},"institutions":[{"id":"https://openalex.org/I114531698","display_name":"Tokyo Institute of Technology","ror":"https://ror.org/0112mx960","country_code":"JP","type":"education","lineage":["https://openalex.org/I114531698"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Manabu Okumura","raw_affiliation_strings":["Tokyo Institute of Technology"],"affiliations":[{"raw_affiliation_string":"Tokyo Institute of Technology","institution_ids":["https://openalex.org/I114531698"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5111702849"],"corresponding_institution_ids":["https://openalex.org/I114531698"],"apc_list":null,"apc_paid":null,"fwci":2.8928,"has_fulltext":true,"cited_by_count":8,"citation_normalized_percentile":{"value":0.91869776,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"3138","last_page":"3153"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9409999847412109,"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":0.9409999847412109,"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":0.7676081657409668},{"id":"https://openalex.org/keywords/distillation","display_name":"Distillation","score":0.739488959312439},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5304056406021118},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.5137806534767151},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4591831862926483},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.33451706171035767},{"id":"https://openalex.org/keywords/chemistry","display_name":"Chemistry","score":0.05995556712150574},{"id":"https://openalex.org/keywords/chromatography","display_name":"Chromatography","score":0.04854917526245117}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7676081657409668},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.739488959312439},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5304056406021118},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5137806534767151},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4591831862926483},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.33451706171035767},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.05995556712150574},{"id":"https://openalex.org/C43617362","wikidata":"https://www.wikidata.org/wiki/Q170050","display_name":"Chromatography","level":1,"score":0.04854917526245117}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2024.findings-naacl.199","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2024.findings-naacl.199","pdf_url":"https://aclanthology.org/2024.findings-naacl.199.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: NAACL 2024","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2024.findings-naacl.199","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2024.findings-naacl.199","pdf_url":"https://aclanthology.org/2024.findings-naacl.199.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: NAACL 2024","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.44999998807907104,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4401043057.pdf","grobid_xml":"https://content.openalex.org/works/W4401043057.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2961085424","https://openalex.org/W4306674287","https://openalex.org/W3046775127","https://openalex.org/W3107602296","https://openalex.org/W4394896187","https://openalex.org/W3170094116","https://openalex.org/W4386462264","https://openalex.org/W4364306694","https://openalex.org/W4312192474","https://openalex.org/W4283697347"],"abstract_inverted_index":{"Dataset":[0],"distillation":[1,37,90],"aims":[2],"to":[3,53,104],"compress":[4],"a":[5,10,44,51,86,101],"training":[6,33,66,108,146],"dataset":[7,36,89,94],"by":[8],"creating":[9],"small":[11],"number":[12],"of":[13,46,50,114,149,154],"informative":[14,106],"synthetic":[15,41,107,117,130],"samples":[16,109],"such":[17,58],"that":[18,128],"neural":[19],"networks":[20],"trained":[21,29],"on":[22,30,121],"them":[23],"perform":[24],"as":[25,27,43,110],"well":[26],"those":[28,135],"the":[31,76],"original":[32],"dataset.Current":[34],"text":[35,52,88,111,123],"methods":[38],"create":[39],"each":[40],"sample":[42],"sequence":[45],"word":[47,70],"embeddings":[48],"instead":[49,113],"apply":[54],"gradient-based":[55],"optimization;":[56],"however,":[57],"embedding-level":[59],"distilled":[60,129],"datasets":[61,125,131],"cannot":[62],"be":[63,160],"used":[64,78],"for":[65,79],"other":[67],"models":[68,150],"whose":[69],"embedding":[71],"weights":[72],"are":[73],"different":[74,147],"from":[75,132,136],"model":[77,103],"distillation.To":[80],"address":[81],"this":[82],"issue,":[83],"we":[84],"propose":[85],"novel":[87],"approach,":[91],"called":[92],"Distilling":[93],"into":[95],"Language":[96],"Model":[97],"(DiLM),":[98],"which":[99],"trains":[100],"language":[102,156],"generate":[105],"data,":[112],"directly":[115],"optimizing":[116],"samples.We":[118],"evaluated":[119],"DiLM":[120,133],"various":[122],"classification":[124],"and":[126,151],"showed":[127],"outperform":[134],"current":[137],"coreset":[138],"selection":[139],"methods.DiLM":[140],"achieved":[141],"remarkable":[142],"generalization":[143],"performance":[144],"in":[145],"types":[148],"in-context":[152],"learning":[153],"large":[155],"models.Our":[157],"code":[158],"will":[159],"available":[161],"at":[162],"https://github.com/arumaekawa/DiLM.":[163]},"counts_by_year":[{"year":2025,"cited_by_count":6},{"year":2024,"cited_by_count":2}],"updated_date":"2026-03-10T14:07:55.174380","created_date":"2025-10-10T00:00:00"}