{"id":"https://openalex.org/W3020987135","doi":"https://doi.org/10.18653/v1/2020.acl-main.357","title":"Pre-training Is (Almost) All You Need: An Application to Commonsense Reasoning","display_name":"Pre-training Is (Almost) All You Need: An Application to Commonsense Reasoning","publication_year":2020,"publication_date":"2020-01-01","ids":{"openalex":"https://openalex.org/W3020987135","doi":"https://doi.org/10.18653/v1/2020.acl-main.357","mag":"3020987135"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2020.acl-main.357","doi":"https://doi.org/10.18653/v1/2020.acl-main.357","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2020.acl-main.357","pdf_url":"https://www.aclweb.org/anthology/2020.acl-main.357.pdf","source":{"id":"https://openalex.org/S4393915740","display_name":"Proceedings of the conference - Association for Computational Linguistics. Meeting","issn_l":"0736-587X","issn":["0736-587X"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["norway-1"],"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 58th Annual Meeting of the Association for Computational Linguistics","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","is_traditional":true,"indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://www.aclweb.org/anthology/2020.acl-main.357.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5017495988","display_name":"Alexandre Tamborrino","orcid":"https://orcid.org/0009-0004-3627-8618","observed_orcids":["https://orcid.org/0009-0004-3627-8618"]},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Alexandre Tamborrino","raw_affiliation_strings":["Samsung Strategy and Innovation Center","[Samsung]"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung Strategy and Innovation Center","institution_ids":["https://openalex.org/I2250650973"]},{"raw_affiliation_string":"[Samsung]","institution_ids":["https://openalex.org/I2250650973"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028421109","display_name":"Nicola Pellican\u00f2","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Nicola Pellican\u00f2","raw_affiliation_strings":["Samsung Strategy and Innovation Center","[Samsung]"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung Strategy and Innovation Center","institution_ids":["https://openalex.org/I2250650973"]},{"raw_affiliation_string":"[Samsung]","institution_ids":["https://openalex.org/I2250650973"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Baptiste Pannier","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Baptiste Pannier","raw_affiliation_strings":["Samsung Strategy and Innovation Center","[Samsung]"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung Strategy and Innovation Center","institution_ids":["https://openalex.org/I2250650973"]},{"raw_affiliation_string":"[Samsung]","institution_ids":["https://openalex.org/I2250650973"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085294714","display_name":"Pascal Voitot","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Pascal Voitot","raw_affiliation_strings":["Samsung Strategy and Innovation Center","[Samsung]"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung Strategy and Innovation Center","institution_ids":["https://openalex.org/I2250650973"]},{"raw_affiliation_string":"[Samsung]","institution_ids":["https://openalex.org/I2250650973"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5080580358","display_name":"Louise Naudin","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Louise Naudin","raw_affiliation_strings":["Samsung Strategy and Innovation Center","[Samsung]"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung Strategy and Innovation Center","institution_ids":["https://openalex.org/I2250650973"]},{"raw_affiliation_string":"[Samsung]","institution_ids":["https://openalex.org/I2250650973"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I2250650973"],"apc_list":null,"apc_paid":null,"fwci":4.159,"has_fulltext":true,"cited_by_count":65,"citation_normalized_percentile":{"value":0.95377588,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":93,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"3878","last_page":"3887"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9599999785423279,"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":0.9599999785423279,"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":0.027400000020861626,"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"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.004600000102072954,"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/commonsense-reasoning","display_name":"commonsense reasoning","score":0.9800000190734863},{"id":"https://openalex.org/keywords/pre-trained-transformers","display_name":"pre-trained transformer models","score":0.6499999761581421},{"id":"https://openalex.org/keywords/masked-language-modeling","display_name":"masked language modeling","score":0.7749999761581421},{"id":"https://openalex.org/keywords/fine-tuning","display_name":"fine-tuning","score":0.7580000162124634},{"id":"https://openalex.org/keywords/hypothesis-ranking","display_name":"hypothesis ranking","score":0.5230000019073486},{"id":"https://openalex.org/keywords/copa","display_name":"COPA","score":0.9390000104904175},{"id":"https://openalex.org/keywords/swag","display_name":"SWAG","score":0.8500000238418579},{"id":"https://openalex.org/keywords/hellaswag","display_name":"HellaSwag","score":0.9980000257492065},{"id":"https://openalex.org/keywords/commonsenseqa","display_name":"CommonsenseQA","score":0.9950000047683716},{"id":"https://openalex.org/keywords/training-stability","display_name":"training stability","score":0.5699999928474426},{"id":"https://openalex.org/keywords/data-efficiency","display_name":"data efficiency","score":0.4869999885559082}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8009267449378967},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.664914608001709},{"id":"https://openalex.org/C95623464","wikidata":"https://www.wikidata.org/wiki/Q1096149","display_name":"Classifier (UML)","level":2,"score":0.6109205484390259},{"id":"https://openalex.org/C193221554","wikidata":"https://www.wikidata.org/wiki/Q5153664","display_name":"Commonsense reasoning","level":2,"score":0.5823479294776917},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5640681982040405},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5636472702026367},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.5078757405281067},{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.45490938425064087},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.41793739795684814},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.37396490573883057},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","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/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.18653/v1/2020.acl-main.357","doi":"https://doi.org/10.18653/v1/2020.acl-main.357","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2020.acl-main.357","pdf_url":"https://www.aclweb.org/anthology/2020.acl-main.357.pdf","source":{"id":"https://openalex.org/S4393915740","display_name":"Proceedings of the conference - Association for Computational Linguistics. Meeting","issn_l":"0736-587X","issn":["0736-587X"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["norway-1"],"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 58th Annual Meeting of the Association for Computational Linguistics","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2004.14074","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2004.14074","pdf_url":"https://arxiv.org/pdf/2004.14074","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":"mag:3020987135","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2004.14074.pdf","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null},{"id":"doi:10.48550/arxiv.2004.14074","doi":"https://doi.org/10.48550/arxiv.2004.14074","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2004.14074","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/v1/2020.acl-main.357","doi":"https://doi.org/10.18653/v1/2020.acl-main.357","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2020.acl-main.357","pdf_url":"https://www.aclweb.org/anthology/2020.acl-main.357.pdf","source":{"id":"https://openalex.org/S4393915740","display_name":"Proceedings of the conference - Association for Computational Linguistics. Meeting","issn_l":"0736-587X","issn":["0736-587X"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["norway-1"],"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 58th Annual Meeting of the Association for Computational Linguistics","raw_type":"proceedings-article"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3020987135.pdf","grobid_xml":"https://content.openalex.org/works/W3020987135.grobid-xml"},"referenced_works_count":22,"referenced_works":["https://openalex.org/W95183648","https://openalex.org/W1599016936","https://openalex.org/W2805206884","https://openalex.org/W2886424491","https://openalex.org/W2898700502","https://openalex.org/W2938704169","https://openalex.org/W2944815030","https://openalex.org/W2945290257","https://openalex.org/W2946609015","https://openalex.org/W2952984539","https://openalex.org/W2963341956","https://openalex.org/W2963995027","https://openalex.org/W2964210047","https://openalex.org/W2965373594","https://openalex.org/W2970476646","https://openalex.org/W2980282514","https://openalex.org/W2981852735","https://openalex.org/W2983915252","https://openalex.org/W2990704537","https://openalex.org/W2996287690","https://openalex.org/W3021191241","https://openalex.org/W3082274269"],"related_works":["https://openalex.org/W4387561502","https://openalex.org/W3196645780","https://openalex.org/W4205553902","https://openalex.org/W3098008462","https://openalex.org/W4293251277","https://openalex.org/W4310429656","https://openalex.org/W3210293355","https://openalex.org/W2973722444","https://openalex.org/W4319985663","https://openalex.org/W2805206884"],"abstract_inverted_index":{"Fine-tuning":[0],"of":[1,19,31,63,107],"pre-trained":[2,43],"transformer":[3],"models":[4],"has":[5,45],"become":[6],"the":[7,20,42,49,64,86,93,101,114,165,199],"standard":[8,186,200],"approach":[9,202],"for":[10],"solving":[11],"common":[12],"NLP":[13],"tasks":[14,99],"(Devlin":[15],"et":[16,117,122,127,133],"al.,":[17,118,123,128,134],"2019).Most":[18],"existing":[21],"approaches":[22],"rely":[23],"on":[24,29,48,113,153,164,189],"a":[25,70,76,81,105,110,175],"randomly":[26],"initialized":[27],"classifier":[28,51,201],"top":[30],"such":[32],"networks.We":[33],"argue":[34],"that":[35,74,155,171],"this":[36,66],"fine-tuning":[37,162],"procedure":[38],"is":[39],"sub-optimal":[40],"as":[41],"model":[44,102],"no":[46],"prior":[47],"specific":[50],"labels,":[52],"while":[53],"it":[54],"might":[55],"have":[56],"already":[57],"learned":[58],"an":[59],"intrinsic":[60],"textual":[61],"representation":[62],"task.In":[65],"paper,":[67],"we":[68,143,169],"introduce":[69],"new":[71],"scoring":[72,139,167],"method":[73,140,173],"casts":[75],"plausibility":[77],"ranking":[78],"task":[79],"in":[80],"full-text":[82],"format":[83],"and":[84,130,193],"leverages":[85],"masked":[87],"language":[88],"modeling":[89],"head":[90],"tuned":[91],"during":[92],"pre-training":[94],"phase.We":[95],"study":[96],"commonsense":[97],"reasoning":[98],"where":[100],"must":[103],"rank":[104],"set":[106],"hypotheses":[108],"given":[109],"premise,":[111],"focusing":[112],"COPA":[115,190],"(Gordon":[116],"2012),":[119],"Swag":[120],"(Zellers":[121,126],"2018),":[124],"HellaSwag":[125],"2019)":[129,135],"CommonsenseQA":[131],"(Talmor":[132],"datasets.By":[136],"exploiting":[137],"our":[138,172],"without":[141],"fine-tuning,":[142],"are":[144,156],"able":[145],"to":[146,158,203],"produce":[147],"strong":[148],"baselines":[149],"(e.g.80%":[150],"test":[151,191],"accuracy":[152],"COPA)":[154],"comparable":[157],"supervised":[159],"approaches.Moreover,":[160],"when":[161],"directly":[163],"proposed":[166],"function,":[168],"show":[170],"provides":[174],"much":[176],"more":[177],"stable":[178],"training":[179],"phase":[180],"across":[181],"random":[182],"restarts":[183],"(e.g":[184],"\u00d710":[185],"deviation":[187],"reduction":[188],"accuracy)":[192],"requires":[194],"less":[195],"annotated":[196],"data":[197],"than":[198],"reach":[204],"equivalent":[205],"performances.":[206]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":11},{"year":2023,"cited_by_count":16},{"year":2022,"cited_by_count":10},{"year":2021,"cited_by_count":22},{"year":2020,"cited_by_count":4}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2020-05-13T00:00:00"}