{"id":"https://openalex.org/W4226063877","doi":"https://doi.org/10.24963/ijcai.2022/383","title":"Human Parity on CommonsenseQA: Augmenting Self-Attention with External Attention","display_name":"Human Parity on CommonsenseQA: Augmenting Self-Attention with External Attention","publication_year":2022,"publication_date":"2022-07-01","ids":{"openalex":"https://openalex.org/W4226063877","doi":"https://doi.org/10.24963/ijcai.2022/383"},"language":"en","primary_location":{"id":"doi:10.24963/ijcai.2022/383","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2022/383","pdf_url":"https://www.ijcai.org/proceedings/2022/0383.pdf","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://www.ijcai.org/proceedings/2022/0383.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5070165835","display_name":"Xu Yi\u2010chong","orcid":"https://orcid.org/0000-0001-6823-8013"},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Yichong Xu","raw_affiliation_strings":["Microsoft"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft","institution_ids":["https://openalex.org/I4210164937"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100355246","display_name":"Chenguang Zhu","orcid":"https://orcid.org/0000-0002-3549-0826"},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Chenguang Zhu","raw_affiliation_strings":["Microsoft"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft","institution_ids":["https://openalex.org/I4210164937"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032047768","display_name":"Shuohang Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Shuohang Wang","raw_affiliation_strings":["Microsoft"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft","institution_ids":["https://openalex.org/I4210164937"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102815325","display_name":"Siqi Sun","orcid":"https://orcid.org/0000-0002-9592-3670"},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Siqi Sun","raw_affiliation_strings":["Microsoft"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft","institution_ids":["https://openalex.org/I4210164937"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100406787","display_name":"Hao Cheng","orcid":"https://orcid.org/0000-0002-3254-4796"},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Hao Cheng","raw_affiliation_strings":["Microsoft"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft","institution_ids":["https://openalex.org/I4210164937"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100374810","display_name":"Xiaodong Liu","orcid":"https://orcid.org/0009-0004-6648-2302"},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Xiaodong Liu","raw_affiliation_strings":["Microsoft Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research","institution_ids":["https://openalex.org/I4210164937"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114910293","display_name":"Jianfeng Gao","orcid":"https://orcid.org/0000-0002-5702-6143"},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Jianfeng Gao","raw_affiliation_strings":["Microsoft Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research","institution_ids":["https://openalex.org/I4210164937"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5019259019","display_name":"Pengcheng He","orcid":"https://orcid.org/0000-0002-7860-503X"},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Pengcheng He","raw_affiliation_strings":["Microsoft"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft","institution_ids":["https://openalex.org/I4210164937"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089195158","display_name":"Michael Zeng","orcid":"https://orcid.org/0000-0001-5302-5883"},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Michael Zeng","raw_affiliation_strings":["Microsoft"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft","institution_ids":["https://openalex.org/I4210164937"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5103158343","display_name":"Xuedong Huang","orcid":"https://orcid.org/0000-0003-4550-7908"},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Xuedong Huang","raw_affiliation_strings":["Microsoft"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft","institution_ids":["https://openalex.org/I4210164937"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210164937"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.03240517,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"2762","last_page":"2768"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.9926999807357788,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.9926999807357788,"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":0.9815999865531921,"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.9815999865531921,"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":0.7178343534469604},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.7087223529815674},{"id":"https://openalex.org/keywords/commonsense-reasoning","display_name":"Commonsense reasoning","score":0.6203588843345642},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5331605672836304},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.4348863959312439},{"id":"https://openalex.org/keywords/democratization","display_name":"Democratization","score":0.4339814782142639},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.10803976655006409},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.0998234748840332}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7178343534469604},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.7087223529815674},{"id":"https://openalex.org/C193221554","wikidata":"https://www.wikidata.org/wiki/Q5153664","display_name":"Commonsense reasoning","level":2,"score":0.6203588843345642},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5331605672836304},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4348863959312439},{"id":"https://openalex.org/C17058734","wikidata":"https://www.wikidata.org/wiki/Q1064441","display_name":"Democratization","level":4,"score":0.4339814782142639},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.10803976655006409},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0998234748840332},{"id":"https://openalex.org/C17744445","wikidata":"https://www.wikidata.org/wiki/Q36442","display_name":"Political science","level":0,"score":0.0},{"id":"https://openalex.org/C555826173","wikidata":"https://www.wikidata.org/wiki/Q7174","display_name":"Democracy","level":3,"score":0.0},{"id":"https://openalex.org/C199539241","wikidata":"https://www.wikidata.org/wiki/Q7748","display_name":"Law","level":1,"score":0.0},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0},{"id":"https://openalex.org/C94625758","wikidata":"https://www.wikidata.org/wiki/Q7163","display_name":"Politics","level":2,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.24963/ijcai.2022/383","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2022/383","pdf_url":"https://www.ijcai.org/proceedings/2022/0383.pdf","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2112.03254","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2112.03254","pdf_url":"https://arxiv.org/pdf/2112.03254","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,"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":"doi:10.48550/arxiv.2112.03254","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2112.03254","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,"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-journal"}],"best_oa_location":{"id":"doi:10.24963/ijcai.2022/383","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2022/383","pdf_url":"https://www.ijcai.org/proceedings/2022/0383.pdf","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4226063877.pdf","grobid_xml":"https://content.openalex.org/works/W4226063877.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W868042","https://openalex.org/W728297","https://openalex.org/W482721","https://openalex.org/W621929","https://openalex.org/W36490","https://openalex.org/W5236451","https://openalex.org/W1160799","https://openalex.org/W14468734","https://openalex.org/W128236","https://openalex.org/W761957"],"abstract_inverted_index":{"Most":[0],"of":[1,16,68,84,108,152,160],"today's":[2],"AI":[3,69,86,94],"systems":[4],"focus":[5,104],"on":[6,13,105,143],"using":[7],"self-attention":[8],"mechanisms":[9],"and":[10,43,64,123],"transformer":[11,32,121],"architectures":[12],"large":[14],"amounts":[15],"diverse":[17,98],"data":[18],"to":[19,29,39,45,57,90,96,156],"achieve":[20],"impressive":[21],"performance":[22,83],"gains.":[23],"In":[24,101],"this":[25],"paper,":[26],"we":[27,55,103],"propose":[28],"augment":[30,119],"the":[31,52,59,66,74,82,106,113,126,144,157],"architecture":[33],"with":[34,149],"an":[35,150],"external":[36,41,49,76,115],"attention":[37,77,116],"mechanism":[38,78,117],"bring":[40],"knowledge":[42],"context":[44],"bear.":[46],"By":[47],"integrating":[48],"information":[50],"into":[51],"prediction":[53],"process,":[54],"hope":[56],"reduce":[58],"need":[60],"for":[61,136],"ever-larger":[62],"models":[63,95,122],"increase":[65],"democratization":[67],"systems.":[70],"We":[71],"find":[72],"that":[73,112],"proposed":[75,114,131],"can":[79,118],"significantly":[80,124],"improve":[81,125],"existing":[85,120],"systems,":[87],"allowing":[88],"practitioners":[89],"easily":[91],"customize":[92],"foundation":[93],"many":[97],"downstream":[99],"applications.":[100],"particular,":[102],"task":[107],"Commonsense":[109],"Reasoning,":[110],"demonstrating":[111],"model's":[127],"reasoning":[128],"capabilities.":[129],"The":[130],"system,":[132],"Knowledgeable":[133],"External":[134],"Attention":[135],"commonsense":[137],"Reasoning":[138],"(KEAR),":[139],"reaches":[140],"human":[141,158],"parity":[142],"open":[145],"CommonsenseQA":[146],"research":[147],"benchmark":[148],"accuracy":[151,159],"89.4%":[153],"in":[154],"comparison":[155],"88.9%.":[161]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2022-05-05T00:00:00"}