{"id":"https://openalex.org/W3152956381","doi":"https://doi.org/10.18653/v1/2021.emnlp-main.243","title":"The Power of Scale for Parameter-Efficient Prompt Tuning","display_name":"The Power of Scale for Parameter-Efficient Prompt Tuning","publication_year":2021,"publication_date":"2021-01-01","ids":{"openalex":"https://openalex.org/W3152956381","doi":"https://doi.org/10.18653/v1/2021.emnlp-main.243","mag":"3152956381"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2021.emnlp-main.243","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2021.emnlp-main.243","pdf_url":"https://aclanthology.org/2021.emnlp-main.243.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"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 2021 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://aclanthology.org/2021.emnlp-main.243.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5021911747","display_name":"Brian Lester","orcid":"https://orcid.org/0000-0002-4297-2508"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Brian Lester","raw_affiliation_strings":["Google Research","Google,,,,,"],"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google,,,,,","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025234902","display_name":"Rami Al\u2010Rfou","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Rami Al-Rfou","raw_affiliation_strings":["Google Research","Google,,,,,"],"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google,,,,,","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5090701958","display_name":"Noah Constant","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Noah Constant","raw_affiliation_strings":["Google Research","Google,,,,,"],"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google,,,,,","institution_ids":["https://openalex.org/I1291425158"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5021911747"],"corresponding_institution_ids":["https://openalex.org/I1291425158"],"apc_list":null,"apc_paid":null,"fwci":10.66791448,"has_fulltext":true,"cited_by_count":94,"citation_normalized_percentile":{"value":0.98925535,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"3045","last_page":"3059"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9998999834060669,"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":0.9998999834060669,"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":0.9994000196456909,"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":0.9965999722480774,"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":0.7976752519607544},{"id":"https://openalex.org/keywords/reuse","display_name":"Reuse","score":0.6946826577186584},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.6718857288360596},{"id":"https://openalex.org/keywords/margin","display_name":"Margin (machine learning)","score":0.6196998953819275},{"id":"https://openalex.org/keywords/fine-tuning","display_name":"Fine-tuning","score":0.5125284194946289},{"id":"https://openalex.org/keywords/downstream","display_name":"Downstream (manufacturing)","score":0.48656976222991943},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.4211674630641937},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3878137469291687},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.35189616680145264}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7976752519607544},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.6946826577186584},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6718857288360596},{"id":"https://openalex.org/C774472","wikidata":"https://www.wikidata.org/wiki/Q6760393","display_name":"Margin (machine learning)","level":2,"score":0.6196998953819275},{"id":"https://openalex.org/C157524613","wikidata":"https://www.wikidata.org/wiki/Q2828883","display_name":"Fine-tuning","level":2,"score":0.5125284194946289},{"id":"https://openalex.org/C2776207758","wikidata":"https://www.wikidata.org/wiki/Q5303302","display_name":"Downstream (manufacturing)","level":2,"score":0.48656976222991943},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4211674630641937},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3878137469291687},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.35189616680145264},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"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/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"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/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0}],"mesh":[],"locations_count":5,"locations":[{"id":"doi:10.18653/v1/2021.emnlp-main.243","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2021.emnlp-main.243","pdf_url":"https://aclanthology.org/2021.emnlp-main.243.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"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 2021 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2104.08691","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2104.08691","pdf_url":"https://arxiv.org/pdf/2104.08691","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":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":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},{"id":"mag:3152956381","is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2104.08691.pdf","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":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":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null},{"id":"doi:10.48550/arxiv.2104.08691","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2104.08691","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":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":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"},{"id":"mag:3212487317","is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2104.08691","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":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":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":"arXiv (Cornell University)","raw_type":null}],"best_oa_location":{"id":"doi:10.18653/v1/2021.emnlp-main.243","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2021.emnlp-main.243","pdf_url":"https://aclanthology.org/2021.emnlp-main.243.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"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 2021 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3152956381.pdf","grobid_xml":"https://content.openalex.org/works/W3152956381.grobid-xml"},"referenced_works_count":50,"referenced_works":["https://openalex.org/W131533222","https://openalex.org/W1599016936","https://openalex.org/W1665214252","https://openalex.org/W2130158090","https://openalex.org/W2135293965","https://openalex.org/W2143612262","https://openalex.org/W2145755360","https://openalex.org/W2396767181","https://openalex.org/W2525127255","https://openalex.org/W2606964149","https://openalex.org/W2746097825","https://openalex.org/W2804897457","https://openalex.org/W2898662126","https://openalex.org/W2919420119","https://openalex.org/W2945290257","https://openalex.org/W2946659172","https://openalex.org/W2950613642","https://openalex.org/W2962739339","https://openalex.org/W2962881743","https://openalex.org/W2962911098","https://openalex.org/W2963026768","https://openalex.org/W2963211188","https://openalex.org/W2963238274","https://openalex.org/W2963250244","https://openalex.org/W2963310665","https://openalex.org/W2963341956","https://openalex.org/W2963403868","https://openalex.org/W2963748441","https://openalex.org/W2963928014","https://openalex.org/W2964303773","https://openalex.org/W2982399380","https://openalex.org/W2988421999","https://openalex.org/W2990704537","https://openalex.org/W2990761674","https://openalex.org/W3006439205","https://openalex.org/W3026404337","https://openalex.org/W3082274269","https://openalex.org/W3098267758","https://openalex.org/W3101498587","https://openalex.org/W3110667630","https://openalex.org/W3111396778","https://openalex.org/W3116459227","https://openalex.org/W3118216348","https://openalex.org/W3128261063","https://openalex.org/W3139080614","https://openalex.org/W3150635270","https://openalex.org/W3153427360","https://openalex.org/W3166846774","https://openalex.org/W3174770825","https://openalex.org/W3174784402"],"related_works":["https://openalex.org/W2963341956","https://openalex.org/W3139080614","https://openalex.org/W3119438769","https://openalex.org/W3082274269","https://openalex.org/W3173777717","https://openalex.org/W3153427360","https://openalex.org/W3098267758","https://openalex.org/W2965373594","https://openalex.org/W3172642864","https://openalex.org/W3098903812","https://openalex.org/W2963403868","https://openalex.org/W3166846774","https://openalex.org/W2964303773","https://openalex.org/W3174784402","https://openalex.org/W2990704537","https://openalex.org/W2963310665","https://openalex.org/W3132736064","https://openalex.org/W3098824823","https://openalex.org/W2970476646","https://openalex.org/W2963846996"],"abstract_inverted_index":{"In":[0],"this":[1,138,165],"work,":[2],"we":[3,74,160,171],"explore":[4],"\"prompt":[5],"tuning\",":[6],"a":[7,62,146,162,175],"simple":[8],"yet":[9],"effective":[10],"mechanism":[11],"for":[12,132],"learning":[13,60],"\"soft":[14],"prompts\"":[15],"to":[16,21,44,120,127,164,185,190],"condition":[17],"frozen":[18,130,176],"language":[19],"models":[20,85,117],"perform":[22],"specific":[23],"downstream":[24,134],"tasks.":[25],"Unlike":[26],"the":[27,93,97,125,149],"discrete":[28],"text":[29],"prompts":[30,35,180],"used":[31],"by":[32,61],"GPT-3,":[33],"soft":[34,179],"are":[36,107,118],"learned":[37,55],"through":[38,67],"backpropagation":[39],"and":[40,95,122,124,156,159,166],"can":[41,136,142],"be":[42,143],"tuned":[43],"incorporate":[45],"signal":[46],"from":[47],"any":[48],"number":[49],"of":[50,88,100,148,154],"labeled":[51],"examples.":[52],"Our":[53,140],"end-to-end":[54],"approach":[56],"outperforms":[57],"GPT-3's":[58],"\"few-shot\"":[59],"large":[63,116],"margin.":[64],"More":[65],"remarkably,":[66],"ablations":[68],"on":[69],"model":[70,101,105,131,177,192],"size":[71],"using":[72],"T5,":[73],"show":[75,172],"that":[76,115,173],"prompt":[77],"tuning":[78,102],"becomes":[79],"more":[80],"competitive":[81],"with":[82,178],"scale:":[83],"as":[84,145,188],"exceed":[86],"billions":[87],"parameters,":[89],"our":[90],"method":[91,141],"\"closes":[92],"gap\"":[94],"matches":[96],"strong":[98],"performance":[99],"(where":[103],"all":[104],"weights":[106],"tuned).":[108],"This":[109],"finding":[110],"is":[111],"especially":[112],"relevant":[113],"in":[114,183],"costly":[119],"share":[121],"serve,":[123],"ability":[126],"reuse":[128],"one":[129],"multiple":[133],"tasks":[135],"ease":[137],"burden.":[139],"seen":[144],"simplification":[147],"recently":[150],"proposed":[151],"\"prefix":[152],"tuning\"":[153],"Li":[155],"Liang":[157],"(2021),":[158],"provide":[161],"comparison":[163],"other":[167],"similar":[168],"approaches.":[169],"Finally,":[170],"conditioning":[174],"confers":[181],"benefits":[182],"robustness":[184],"domain":[186],"transfer,":[187],"compared":[189],"full":[191],"tuning.":[193]},"counts_by_year":[{"year":2025,"cited_by_count":5},{"year":2024,"cited_by_count":12},{"year":2023,"cited_by_count":7},{"year":2022,"cited_by_count":22},{"year":2021,"cited_by_count":46},{"year":2020,"cited_by_count":2}],"updated_date":"2026-02-09T09:26:11.010843","created_date":"2025-10-10T00:00:00"}