{"id":"https://openalex.org/W3200092677","doi":"https://doi.org/10.1109/ijcnn52387.2021.9534299","title":"How Can the [MASK] Know? The Sources and Limitations of Knowledge in BERT","display_name":"How Can the [MASK] Know? The Sources and Limitations of Knowledge in BERT","publication_year":2021,"publication_date":"2021-07-18","ids":{"openalex":"https://openalex.org/W3200092677","doi":"https://doi.org/10.1109/ijcnn52387.2021.9534299","mag":"3200092677"},"language":"en","primary_location":{"id":"doi:10.1109/ijcnn52387.2021.9534299","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ijcnn52387.2021.9534299","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 International Joint Conference on Neural Networks (IJCNN)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5039991534","display_name":"Maksim Podkorytov","orcid":"https://orcid.org/0000-0003-3312-8038"},"institutions":[{"id":"https://openalex.org/I103163165","display_name":"Florida State University","ror":"https://ror.org/05g3dte14","country_code":"US","type":"education","lineage":["https://openalex.org/I103163165"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Maksim Podkorytov","raw_affiliation_strings":["Florida State University, Tallahassee, FL, USA"],"affiliations":[{"raw_affiliation_string":"Florida State University, Tallahassee, FL, USA","institution_ids":["https://openalex.org/I103163165"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008250482","display_name":"Daniel Bi\u015b","orcid":"https://orcid.org/0000-0003-1347-0151"},"institutions":[{"id":"https://openalex.org/I103163165","display_name":"Florida State University","ror":"https://ror.org/05g3dte14","country_code":"US","type":"education","lineage":["https://openalex.org/I103163165"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Daniel Bis","raw_affiliation_strings":["Florida State University, Tallahassee, FL, USA"],"affiliations":[{"raw_affiliation_string":"Florida State University, Tallahassee, FL, USA","institution_ids":["https://openalex.org/I103163165"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5102867647","display_name":"Xiuwen Liu","orcid":"https://orcid.org/0000-0002-9320-3872"},"institutions":[{"id":"https://openalex.org/I103163165","display_name":"Florida State University","ror":"https://ror.org/05g3dte14","country_code":"US","type":"education","lineage":["https://openalex.org/I103163165"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xiuwen Liu","raw_affiliation_strings":["Florida State University, Tallahassee, FL, USA"],"affiliations":[{"raw_affiliation_string":"Florida State University, Tallahassee, FL, USA","institution_ids":["https://openalex.org/I103163165"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5039991534"],"corresponding_institution_ids":["https://openalex.org/I103163165"],"apc_list":null,"apc_paid":null,"fwci":1.7676,"has_fulltext":false,"cited_by_count":15,"citation_normalized_percentile":{"value":0.87609511,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"8"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":1.0,"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":1.0,"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":0.9980999827384949,"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":0.9925000071525574,"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":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7677682638168335},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6115890145301819},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.5752753019332886},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.5598928928375244},{"id":"https://openalex.org/keywords/transferability","display_name":"Transferability","score":0.5475775003433228},{"id":"https://openalex.org/keywords/point","display_name":"Point (geometry)","score":0.5355152487754822},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5077217817306519},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.4916563630104065},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.44023963809013367},{"id":"https://openalex.org/keywords/confusion","display_name":"Confusion","score":0.43642476201057434},{"id":"https://openalex.org/keywords/downstream","display_name":"Downstream (manufacturing)","score":0.42500150203704834},{"id":"https://openalex.org/keywords/question-answering","display_name":"Question answering","score":0.4239976108074188},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.41626495122909546},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.3603006601333618},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.29033753275871277},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.18980073928833008},{"id":"https://openalex.org/keywords/epistemology","display_name":"Epistemology","score":0.11995518207550049},{"id":"https://openalex.org/keywords/psychology","display_name":"Psychology","score":0.09019345045089722}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7677682638168335},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6115890145301819},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.5752753019332886},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5598928928375244},{"id":"https://openalex.org/C61272859","wikidata":"https://www.wikidata.org/wiki/Q7834031","display_name":"Transferability","level":3,"score":0.5475775003433228},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.5355152487754822},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5077217817306519},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.4916563630104065},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.44023963809013367},{"id":"https://openalex.org/C2781140086","wikidata":"https://www.wikidata.org/wiki/Q557945","display_name":"Confusion","level":2,"score":0.43642476201057434},{"id":"https://openalex.org/C2776207758","wikidata":"https://www.wikidata.org/wiki/Q5303302","display_name":"Downstream (manufacturing)","level":2,"score":0.42500150203704834},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.4239976108074188},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.41626495122909546},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.3603006601333618},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.29033753275871277},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.18980073928833008},{"id":"https://openalex.org/C111472728","wikidata":"https://www.wikidata.org/wiki/Q9471","display_name":"Epistemology","level":1,"score":0.11995518207550049},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.09019345045089722},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","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/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","level":1,"score":0.0},{"id":"https://openalex.org/C140331021","wikidata":"https://www.wikidata.org/wiki/Q1868104","display_name":"Logit","level":2,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C11171543","wikidata":"https://www.wikidata.org/wiki/Q41630","display_name":"Psychoanalysis","level":1,"score":0.0},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/ijcnn52387.2021.9534299","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ijcnn52387.2021.9534299","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 International Joint Conference on Neural Networks (IJCNN)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.7799999713897705,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":37,"referenced_works":["https://openalex.org/W1647671624","https://openalex.org/W2042528187","https://openalex.org/W2078483536","https://openalex.org/W2525778437","https://openalex.org/W2896457183","https://openalex.org/W2899663614","https://openalex.org/W2910243263","https://openalex.org/W2930944935","https://openalex.org/W2950339735","https://openalex.org/W2962739339","https://openalex.org/W2963341956","https://openalex.org/W2963403868","https://openalex.org/W2964110616","https://openalex.org/W2965373594","https://openalex.org/W2970476646","https://openalex.org/W2970597249","https://openalex.org/W2970986510","https://openalex.org/W2971869958","https://openalex.org/W2972167903","https://openalex.org/W2972324944","https://openalex.org/W2977944219","https://openalex.org/W2980282514","https://openalex.org/W2995446988","https://openalex.org/W2996428491","https://openalex.org/W3007672467","https://openalex.org/W3021533447","https://openalex.org/W3033176962","https://openalex.org/W3044438666","https://openalex.org/W3048920610","https://openalex.org/W3100754494","https://openalex.org/W3111372685","https://openalex.org/W3118485687","https://openalex.org/W4287824654","https://openalex.org/W4385245566","https://openalex.org/W6636915900","https://openalex.org/W6739901393","https://openalex.org/W6755207826"],"related_works":["https://openalex.org/W2161221533","https://openalex.org/W4229699405","https://openalex.org/W1666484574","https://openalex.org/W2216382288","https://openalex.org/W2355491300","https://openalex.org/W4234629551","https://openalex.org/W3204607391","https://openalex.org/W2964413124","https://openalex.org/W4388937922","https://openalex.org/W3113264705"],"abstract_inverted_index":{"We":[0],"explore":[1],"the":[2,6,19,34,37,46,63,68,87,102,105,114,119,130,176],"idea":[3],"of":[4,12,18,36,58,113,175],"using":[5,142],"pre-trained":[7],"BERT":[8,100,150,168],"as":[9],"a":[10,143],"source":[11],"factual":[13,30],"knowledge,":[14,31],"analyze":[15],"which":[16,156],"components":[17],"model":[20,64],"are":[21,182],"responsible":[22],"for":[23,52],"its":[24],"ability":[25],"to":[26,39,71,85,101,153],"answer":[27,106],"questions":[28],"requiring":[29],"and":[32,124,171],"study":[33],"transferability":[35,57],"knowledge":[38,60,74,121],"downstream":[40,162],"tasks.":[41],"Our":[42],"experiments":[43,181],"show":[44,148],"that":[45,56,98,104,118,149,167],"Language":[47],"Modeling":[48],"Head":[49],"is":[50,65,75,122,151],"indispensable":[51],"predicting":[53],"facts,":[54],"implying":[55],"any":[59],"captured":[61],"in":[62,77,129],"limited.":[66],"While":[67],"dominant":[69],"approach":[70],"researching":[72],"how":[73],"stored":[76,120],"language":[78],"models":[79],"focuses":[80],"on":[81,126,161,184],"tailoring":[82],"question":[83,92],"formulation":[84],"optimize":[86],"retrieval":[88],"quality,":[89],"we":[90,147,165],"find":[91],"patterns":[93,116],"easily":[94],"understood":[95],"by":[96],"humans":[97],"confuse":[99],"point":[103],"does":[107],"not":[108],"make":[109],"sense.":[110],"The":[111],"nature":[112],"found":[115],"implies":[117],"fragile":[123],"based":[125],"token":[127],"co-occurrence":[128],"training":[131],"set":[132],"used":[133],"during":[134],"pre-training,":[135],"rather":[136],"than":[137],"generalization":[138],"or":[139],"inference.":[140],"Moreover,":[141],"novel,":[144],"hand-crafted":[145],"dataset":[146],"vulnerable":[152],"common":[154],"misconceptions,":[155],"could":[157],"have":[158],"fatal":[159],"effects":[160],"applications.":[163],"Overall,":[164],"conclude":[166],"offers":[169],"low":[170],"unreliable":[172],"performance":[173],"out":[174],"box.":[177],"Jupyter":[178],"notebooks":[179],"with":[180],"available":[183],"GitHub.":[185],"<sup":[186,189],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[187,190],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">1</sup>":[188,191],"Project":[192],"URL:":[193],"https://github.com/tenpercent/knowledge-and-confusion":[194]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":3},{"year":2023,"cited_by_count":8},{"year":2022,"cited_by_count":2}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}