{"id":"https://openalex.org/W4389524481","doi":"https://doi.org/10.18653/v1/2023.emnlp-main.189","title":"Lion: Adversarial Distillation of Proprietary Large Language Models","display_name":"Lion: Adversarial Distillation of Proprietary Large Language Models","publication_year":2023,"publication_date":"2023-01-01","ids":{"openalex":"https://openalex.org/W4389524481","doi":"https://doi.org/10.18653/v1/2023.emnlp-main.189"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2023.emnlp-main.189","doi":"https://doi.org/10.18653/v1/2023.emnlp-main.189","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2023.emnlp-main.189","pdf_url":"https://aclanthology.org/2023.emnlp-main.189.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Conference on Empirical Methods in Natural Language Processing (EMNLP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://aclanthology.org/2023.emnlp-main.189.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5122087483","display_name":"Yuxin Jiang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]},{"id":"https://openalex.org/I4405261251","display_name":"The Hong Kong University of Science and Technology (Guangzhou)","ror":"https://ror.org/050h0vm43","country_code":"CN","type":"education","lineage":["https://openalex.org/I4405261251"]}],"countries":["CN","HK"],"is_corresponding":false,"raw_author_name":"Yuxin Jiang","raw_affiliation_strings":["The Hong Kong University of Science and Technology (Guangzhou), Guangzhou, China","The Hong Kong University of Science and Technology, Hong Kong SAR, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology (Guangzhou), Guangzhou, China","institution_ids":["https://openalex.org/I4405261251"]},{"raw_affiliation_string":"The Hong Kong University of Science and Technology, Hong Kong SAR, China","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018499468","display_name":"Chunkit Chan","orcid":"https://orcid.org/0000-0002-1520-4597","observed_orcids":["https://orcid.org/0000-0002-1520-4597"]},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Chunkit Chan","raw_affiliation_strings":["The Hong Kong University of Science and Technology, Hong Kong SAR, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology, Hong Kong SAR, China","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Mingyang Chen","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]},{"id":"https://openalex.org/I4405261251","display_name":"The Hong Kong University of Science and Technology (Guangzhou)","ror":"https://ror.org/050h0vm43","country_code":"CN","type":"education","lineage":["https://openalex.org/I4405261251"]}],"countries":["CN","HK"],"is_corresponding":false,"raw_author_name":"Mingyang Chen","raw_affiliation_strings":["The Hong Kong University of Science and Technology (Guangzhou), Guangzhou, China","The Hong Kong University of Science and Technology, Hong Kong SAR, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology (Guangzhou), Guangzhou, China","institution_ids":["https://openalex.org/I4405261251"]},{"raw_affiliation_string":"The Hong Kong University of Science and Technology, Hong Kong SAR, China","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100391888","display_name":"Wei Wang","orcid":"https://orcid.org/0000-0002-1568-2396","observed_orcids":["https://orcid.org/0000-0002-1568-2396","https://orcid.org/0000-0002-2908-3060","https://orcid.org/0000-0002-2772-4856","https://orcid.org/0009-0000-8893-8375","https://orcid.org/0000-0003-0264-788X","https://orcid.org/0000-0002-1717-5785","https://orcid.org/0000-0002-8180-2886","https://orcid.org/0000-0001-7608-7438","https://orcid.org/0000-0003-2262-2508"]},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]},{"id":"https://openalex.org/I4405261251","display_name":"The Hong Kong University of Science and Technology (Guangzhou)","ror":"https://ror.org/050h0vm43","country_code":"CN","type":"education","lineage":["https://openalex.org/I4405261251"]}],"countries":["CN","HK"],"is_corresponding":false,"raw_author_name":"Wei Wang","raw_affiliation_strings":["The Hong Kong University of Science and Technology (Guangzhou), Guangzhou, China","The Hong Kong University of Science and Technology, Hong Kong SAR, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology (Guangzhou), Guangzhou, China","institution_ids":["https://openalex.org/I4405261251"]},{"raw_affiliation_string":"The Hong Kong University of Science and Technology, Hong Kong SAR, China","institution_ids":["https://openalex.org/I200769079"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":6.3632,"has_fulltext":true,"cited_by_count":37,"citation_normalized_percentile":{"value":0.97594988,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":89,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"3134","last_page":"3154"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.8101000189781189,"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":0.8101000189781189,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.11599999666213989,"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":0.037700001150369644,"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/knowledge-distillation","display_name":"knowledge distillation","score":0.878000020980835},{"id":"https://openalex.org/keywords/adversarial-distillation","display_name":"adversarial distillation","score":0.8339999914169312},{"id":"https://openalex.org/keywords/open-source-llms","display_name":"open-source LLMs","score":0.7200000286102295},{"id":"https://openalex.org/keywords/chatgpt","display_name":"ChatGPT","score":0.8410000205039978},{"id":"https://openalex.org/keywords/zero-shot-reasoning","display_name":"zero-shot reasoning","score":0.8180000185966492},{"id":"https://openalex.org/keywords/open-ended-generation","display_name":"open-ended generation","score":0.6359999775886536},{"id":"https://openalex.org/keywords/instruction-tuning","display_name":"instruction tuning","score":0.3790000081062317}],"concepts":[{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.8948179483413696},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7990052700042725},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.6055047512054443},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5395665764808655},{"id":"https://openalex.org/C126388530","wikidata":"https://www.wikidata.org/wiki/Q1131737","display_name":"Imitation","level":2,"score":0.5206655263900757},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5200247764587402},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.501969575881958},{"id":"https://openalex.org/C177606310","wikidata":"https://www.wikidata.org/wiki/Q5674297","display_name":"Adaptability","level":2,"score":0.415388822555542},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.35266149044036865},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.22592973709106445},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C178790620","wikidata":"https://www.wikidata.org/wiki/Q11351","display_name":"Organic chemistry","level":1,"score":0.0},{"id":"https://openalex.org/C18903297","wikidata":"https://www.wikidata.org/wiki/Q7150","display_name":"Ecology","level":1,"score":0.0},{"id":"https://openalex.org/C77805123","wikidata":"https://www.wikidata.org/wiki/Q161272","display_name":"Social psychology","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.18653/v1/2023.emnlp-main.189","doi":"https://doi.org/10.18653/v1/2023.emnlp-main.189","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2023.emnlp-main.189","pdf_url":"https://aclanthology.org/2023.emnlp-main.189.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Conference on Empirical Methods in Natural Language Processing (EMNLP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2305.12870","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2305.12870","pdf_url":"https://arxiv.org/pdf/2305.12870","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"pmh:oai:repository.hkust.edu.hk:1783.1-136390","doi":null,"is_oa":false,"landing_page_url":"http://lbdiscover.ust.hk/uresolver?url_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rfr_id=info:sid/HKUST:SPI&rft.genre=article&rft.issn=&rft.volume=&rft.issue=&rft.date=2023&rft.spage=&rft.aulast=Jiang&rft.aufirst=Yuxin&rft.atitle=Lion%3A+Adversarial+Distillation+of+Proprietary+Large+Language+Models&rft.title=","pdf_url":null,"source":{"id":"https://openalex.org/S4306401796","display_name":"Rare & Special e-Zone (The Hong Kong University of Science and Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I200769079","host_organization_name":"Hong Kong University of Science and Technology","host_organization_lineage":["https://openalex.org/I200769079"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Conference paper"},{"id":"doi:10.48550/arxiv.2305.12870","doi":"https://doi.org/10.48550/arxiv.2305.12870","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2305.12870","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.18653/v1/2023.emnlp-main.189","doi":"https://doi.org/10.18653/v1/2023.emnlp-main.189","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2023.emnlp-main.189","pdf_url":"https://aclanthology.org/2023.emnlp-main.189.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Conference on Empirical Methods in Natural Language Processing (EMNLP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"sustainable_development_goals":[{"display_name":"Quality education","score":0.4156000018119812,"id":"https://metadata.un.org/sdg/4"}],"study_designs":[],"awards":[{"id":"https://openalex.org/G7020882947","display_name":null,"funder_award_id":"2023A03J0003","funder_id":"https://openalex.org/F4320335480","funder_display_name":"Guangzhou Municipal Science and Technology Project"}],"funders":[{"id":"https://openalex.org/F4320323537","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597"},{"id":"https://openalex.org/F4320335480","display_name":"Guangzhou Municipal Science and Technology Project","ror":null}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4389524481.pdf","grobid_xml":"https://content.openalex.org/works/W4389524481.grobid-xml"},"referenced_works_count":41,"referenced_works":["https://openalex.org/W1821462560","https://openalex.org/W2603766943","https://openalex.org/W2905209730","https://openalex.org/W2946348289","https://openalex.org/W2963785012","https://openalex.org/W2964016283","https://openalex.org/W2981852735","https://openalex.org/W2996610304","https://openalex.org/W2996834813","https://openalex.org/W3023663521","https://openalex.org/W3034957837","https://openalex.org/W3035317912","https://openalex.org/W3109310454","https://openalex.org/W3119373805","https://openalex.org/W3196731672","https://openalex.org/W3198002980","https://openalex.org/W3205068155","https://openalex.org/W3217187998","https://openalex.org/W4221143046","https://openalex.org/W4226278401","https://openalex.org/W4281690148","https://openalex.org/W4283026156","https://openalex.org/W4283816018","https://openalex.org/W4306808908","https://openalex.org/W4312091890","https://openalex.org/W4319793302","https://openalex.org/W4322718191","https://openalex.org/W4360836968","https://openalex.org/W4367000491","https://openalex.org/W4367628242","https://openalex.org/W4378499145","https://openalex.org/W4378770815","https://openalex.org/W4379539933","https://openalex.org/W4385570578","https://openalex.org/W4385573468","https://openalex.org/W4386510385","https://openalex.org/W4386841414","https://openalex.org/W4387725212","https://openalex.org/W4389518968","https://openalex.org/W4389523788","https://openalex.org/W4401042689"],"related_works":["https://openalex.org/W7221113101","https://openalex.org/W4380993498","https://openalex.org/W7128796328","https://openalex.org/W4416034701","https://openalex.org/W7131214503","https://openalex.org/W4404306953","https://openalex.org/W7065441497","https://openalex.org/W4416454742","https://openalex.org/W7221123345","https://openalex.org/W4404782133"],"abstract_inverted_index":{"The":[0],"practice":[1],"of":[2,35,41,48,54,92,117],"transferring":[3],"knowledge":[4,28,86,129],"from":[5,130],"a":[6,14,26,46,77,83,113,133,139],"sophisticated,":[7],"proprietary":[8],"large":[9],"language":[10],"model":[11,38,98,135],"(LLM)":[12],"to":[13,45,99,132,155],"compact,":[15],"open-source":[16],"LLM":[17],"has":[18],"garnered":[19],"considerable":[20],"attention.Previous":[21],"works":[22],"have":[23],"focused":[24],"on":[25,180],"unidirectional":[27],"distillation":[29,80],"way":[30],"by":[31,166],"aligning":[32],"the":[33,36,42,52,61,68,88,96,109],"responses":[34],"student":[37,62,69,110,134],"with":[39],"those":[40],"teacher":[43,97],"models":[44,163],"set":[47],"instructions.Nevertheless,":[49],"they":[50],"overlooked":[51],"possibility":[53],"incorporating":[55],"any":[56],"\"feedback\"-identifying":[57],"challenging":[58,169],"instructions":[59,102,107],"where":[60],"model's":[63,70],"performance":[64],"falls":[65],"short-to":[66],"boost":[67],"proficiency":[71],"iteratively.To":[72],"this":[73,123],"end,":[74],"we":[75,94,126],"propose":[76],"novel":[78],"adversarial":[79,115,124],"framework":[81],"for":[82,108],"more":[84],"efficient":[85],"transfer.Leveraging":[87],"versatile":[89],"role":[90],"adaptability":[91],"LLMs,":[93],"prompt":[95],"identify":[100],"\"hard\"":[101,106],"and":[103,120,178],"generate":[104],"new":[105],"model,":[111],"creating":[112],"three-stage":[114],"loop":[116],"imitation,":[118],"discrimination,":[119],"generation.By":[121],"applying":[122],"framework,":[125],"successfully":[127],"transfer":[128],"ChatGPT":[131],"(named":[136],"Lion),":[137],"using":[138],"mere":[140],"70k":[141],"training":[142],"data.Our":[143],"results":[144],"show":[145],"that":[146],"Lion-13B":[147],"not":[148],"only":[149],"achieves":[150],"comparable":[151],"open-ended":[152],"generation":[153],"capabilities":[154],"Chat-GPT":[156],"but":[157],"surpasses":[158],"conventional":[159],"state-of-the-art":[160],"(SOTA)":[161],"instruction-tuned":[162],"like":[164],"Vicuna-13B":[165],"55.4%":[167],"in":[168],"zero-shot":[170],"reasoning":[171],"benchmarks":[172],"such":[173],"as":[174],"BIG-Bench":[175],"Hard":[176],"(BBH)":[177],"16.7%":[179],"AGIEval.":[181],"1":[182]},"counts_by_year":[{"year":2026,"cited_by_count":6},{"year":2025,"cited_by_count":14},{"year":2024,"cited_by_count":11},{"year":2023,"cited_by_count":5},{"year":2022,"cited_by_count":1}],"updated_date":"2026-10-30T12:39:25.321706","created_date":"2023-12-11T00:00:00"}