{"id":"https://openalex.org/W4392012666","doi":"https://doi.org/10.48550/arxiv.2402.12842","title":"PromptKD: Distilling Student-Friendly Knowledge for Generative Language\n Models via Prompt Tuning","display_name":"PromptKD: Distilling Student-Friendly Knowledge for Generative Language\n Models via Prompt Tuning","publication_year":2024,"publication_date":"2024-02-20","ids":{"openalex":"https://openalex.org/W4392012666","doi":"https://doi.org/10.48550/arxiv.2402.12842"},"language":"en","primary_location":{"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2402.12842","pdf_url":"http://arxiv.org/pdf/2402.12842","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_indexed_in_scopus":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":["Cornell University"],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false},"type":"preprint","type_crossref":"posted-content","indexed_in":["arxiv"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"http://arxiv.org/pdf/2402.12842","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5040969166","display_name":"Gyeongman Kim","orcid":"https://orcid.org/0009-0004-6281-9203"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kim, Gyeongman","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102620279","display_name":"Doohyuk Jang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jang, Doohyuk","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5086698569","display_name":"Eunho Yang","orcid":"https://orcid.org/0000-0003-2188-0169"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Eunho","raw_affiliation_strings":[],"affiliations":[]}],"institution_assertions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.0,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":0,"max":77},"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9993,"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":0.9993,"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/T10028","display_name":"Topic Modeling","score":0.9983,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9822,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[],"concepts":[{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.7974242},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.59017706},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.33006155}],"mesh":[],"locations_count":1,"locations":[{"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2402.12842","pdf_url":"http://arxiv.org/pdf/2402.12842","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_indexed_in_scopus":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":["Cornell University"],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false}],"best_oa_location":{"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2402.12842","pdf_url":"http://arxiv.org/pdf/2402.12842","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_indexed_in_scopus":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":["Cornell University"],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false},"sustainable_development_goals":[],"grants":[],"datasets":[],"versions":[],"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391913857","https://openalex.org/W2748952813","https://openalex.org/W2478288626","https://openalex.org/W2390279801","https://openalex.org/W2382290278","https://openalex.org/W2380075625","https://openalex.org/W2376932109","https://openalex.org/W2358668433","https://openalex.org/W2350741829","https://openalex.org/W2001405890"],"abstract_inverted_index":{"Recent":[0],"advancements":[1],"in":[2,55,62,87,102],"large":[3],"language":[4,36,64,93],"models":[5,37,94],"(LLMs)":[6],"have":[7],"raised":[8],"concerns":[9],"about":[10],"inference":[11],"costs,":[12],"increasing":[13],"the":[14,44,84,107,130,141,156,173],"need":[15],"for":[16,29,34,57,83,111],"research":[17,31],"into":[18],"model":[19,110,143],"compression.":[20],"While":[21],"knowledge":[22,167],"distillation":[23],"(KD)":[24],"is":[25,40],"a":[26,73,121],"prominent":[27],"method":[28,77],"this,":[30],"on":[32,137],"KD":[33,56,88],"generative":[35,63,92],"like":[38],"LLMs":[39],"relatively":[41],"sparse,":[42],"and":[43,127],"approach":[45],"of":[46,124,155],"distilling":[47,165],"student-friendly":[48,97,113,166],"knowledge,":[49,114],"which":[50],"has":[51],"shown":[52],"promising":[53],"performance":[54,150,179],"classification":[58,103],"models,":[59],"remains":[60],"unexplored":[61],"models.":[65],"To":[66],"explore":[67],"this":[68],"approach,":[69],"we":[70],"propose":[71],"PromptKD,":[72],"simple":[74],"yet":[75],"effective":[76],"that":[78,104,146,164],"utilizes":[79],"prompt":[80,125,131],"tuning":[81,128],"-":[82,89],"first":[85],"time":[86],"to":[90,95,178],"enable":[91],"transfer":[96],"knowledge.":[98],"Unlike":[99],"previous":[100],"works":[101],"require":[105],"fine-tuning":[106],"entire":[108,174],"teacher":[109],"extracting":[112],"PromptKD":[115,147],"achieves":[116,148],"similar":[117],"effects":[118],"by":[119],"adding":[120,152],"small":[122],"number":[123],"tokens":[126],"only":[129,153],"with":[132],"student":[133],"guidance.":[134],"Extensive":[135],"experiments":[136],"instruction-following":[138],"datasets":[139],"using":[140],"GPT-2":[142],"family":[144],"show":[145],"state-of-the-art":[149],"while":[151],"0.0007%":[154],"teacher's":[157],"parameters":[158],"as":[159],"prompts.":[160],"Further":[161],"analysis":[162],"suggests":[163],"alleviates":[168],"exposure":[169],"bias":[170],"effectively":[171],"throughout":[172],"training":[175],"process,":[176],"leading":[177],"enhancements.":[180]},"abstract_inverted_index_v3":null,"cited_by_api_url":"https://api.openalex.org/works?filter=cites:W4392012666","counts_by_year":[],"updated_date":"2025-04-21T05:05:09.299737","created_date":"2024-02-22"}