{"id":"https://openalex.org/W3006188107","doi":"https://doi.org/10.24963/ijcai.2020/537","title":"Transformers as Soft Reasoners over Language","display_name":"Transformers as Soft Reasoners over Language","publication_year":2020,"publication_date":"2020-07-01","ids":{"openalex":"https://openalex.org/W3006188107","doi":"https://doi.org/10.24963/ijcai.2020/537","mag":"3006188107"},"language":"en","primary_location":{"id":"doi:10.24963/ijcai.2020/537","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2020/537","pdf_url":"https://www.ijcai.org/proceedings/2020/0537.pdf","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://www.ijcai.org/proceedings/2020/0537.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5081635691","display_name":"Peter E. Clark","orcid":"https://orcid.org/0000-0002-8006-7015"},"institutions":[{"id":"https://openalex.org/I4210140341","display_name":"Allen Institute","ror":"https://ror.org/03cpe7c52","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I4210140341"]},{"id":"https://openalex.org/I4210156221","display_name":"Allen Institute for Artificial Intelligence","ror":"https://ror.org/05w520734","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I4210156221"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Peter Clark","raw_affiliation_strings":["Allen Institute for AI","ALLEN INSTITUTE FOR ARTIFICIAL INTELLIGENCE"],"affiliations":[{"raw_affiliation_string":"Allen Institute for AI","institution_ids":["https://openalex.org/I4210140341"]},{"raw_affiliation_string":"ALLEN INSTITUTE FOR ARTIFICIAL INTELLIGENCE","institution_ids":["https://openalex.org/I4210156221"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071250061","display_name":"Oyvind Tafjord","orcid":null},"institutions":[{"id":"https://openalex.org/I4210140341","display_name":"Allen Institute","ror":"https://ror.org/03cpe7c52","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I4210140341"]},{"id":"https://openalex.org/I4210156221","display_name":"Allen Institute for Artificial Intelligence","ror":"https://ror.org/05w520734","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I4210156221"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Oyvind Tafjord","raw_affiliation_strings":["Allen Institute for AI","ALLEN INSTITUTE FOR ARTIFICIAL INTELLIGENCE"],"affiliations":[{"raw_affiliation_string":"Allen Institute for AI","institution_ids":["https://openalex.org/I4210140341"]},{"raw_affiliation_string":"ALLEN INSTITUTE FOR ARTIFICIAL INTELLIGENCE","institution_ids":["https://openalex.org/I4210156221"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5039359682","display_name":"Kyle Richardson","orcid":"https://orcid.org/0000-0003-4836-0753"},"institutions":[{"id":"https://openalex.org/I4210156221","display_name":"Allen Institute for Artificial Intelligence","ror":"https://ror.org/05w520734","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I4210156221"]},{"id":"https://openalex.org/I4210140341","display_name":"Allen Institute","ror":"https://ror.org/03cpe7c52","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I4210140341"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kyle Richardson","raw_affiliation_strings":["Allen Institute for AI","ALLEN INSTITUTE FOR ARTIFICIAL INTELLIGENCE"],"affiliations":[{"raw_affiliation_string":"Allen Institute for AI","institution_ids":["https://openalex.org/I4210140341"]},{"raw_affiliation_string":"ALLEN INSTITUTE FOR ARTIFICIAL INTELLIGENCE","institution_ids":["https://openalex.org/I4210156221"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5081635691"],"corresponding_institution_ids":["https://openalex.org/I4210140341","https://openalex.org/I4210156221"],"apc_list":null,"apc_paid":null,"fwci":2.34975278,"has_fulltext":false,"cited_by_count":16,"citation_normalized_percentile":{"value":0.89573893,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"3882","last_page":"3890"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9997000098228455,"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":0.9997000098228455,"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":0.9990000128746033,"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/T11303","display_name":"Bayesian Modeling and Causal Inference","score":0.9947999715805054,"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":0.7539524435997009},{"id":"https://openalex.org/keywords/counterfactual-thinking","display_name":"Counterfactual thinking","score":0.6525524258613586},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.6114497184753418},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5865216851234436},{"id":"https://openalex.org/keywords/forward-chaining","display_name":"Forward chaining","score":0.519076406955719},{"id":"https://openalex.org/keywords/question-answering","display_name":"Question answering","score":0.5101625919342041},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.49409249424934387},{"id":"https://openalex.org/keywords/knowledge-representation-and-reasoning","display_name":"Knowledge representation and reasoning","score":0.48652806878089905},{"id":"https://openalex.org/keywords/natural-language-understanding","display_name":"Natural language understanding","score":0.477181077003479},{"id":"https://openalex.org/keywords/probabilistic-logic","display_name":"Probabilistic logic","score":0.4480288326740265},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.40030765533447266},{"id":"https://openalex.org/keywords/expert-system","display_name":"Expert system","score":0.27703213691711426},{"id":"https://openalex.org/keywords/psychology","display_name":"Psychology","score":0.11540943384170532}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7539524435997009},{"id":"https://openalex.org/C108650721","wikidata":"https://www.wikidata.org/wiki/Q1783253","display_name":"Counterfactual thinking","level":2,"score":0.6525524258613586},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.6114497184753418},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5865216851234436},{"id":"https://openalex.org/C142614401","wikidata":"https://www.wikidata.org/wiki/Q777433","display_name":"Forward chaining","level":3,"score":0.519076406955719},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.5101625919342041},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.49409249424934387},{"id":"https://openalex.org/C161301231","wikidata":"https://www.wikidata.org/wiki/Q3478658","display_name":"Knowledge representation and reasoning","level":2,"score":0.48652806878089905},{"id":"https://openalex.org/C2779439875","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Natural language understanding","level":3,"score":0.477181077003479},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.4480288326740265},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.40030765533447266},{"id":"https://openalex.org/C58328972","wikidata":"https://www.wikidata.org/wiki/Q184609","display_name":"Expert system","level":2,"score":0.27703213691711426},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.11540943384170532},{"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},{"id":"https://openalex.org/C77805123","wikidata":"https://www.wikidata.org/wiki/Q161272","display_name":"Social psychology","level":1,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.24963/ijcai.2020/537","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2020/537","pdf_url":"https://www.ijcai.org/proceedings/2020/0537.pdf","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2002.05867","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2002.05867","pdf_url":"https://arxiv.org/pdf/2002.05867","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:3006188107","is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2002.05867.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.2002.05867","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2002.05867","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"}],"best_oa_location":{"id":"doi:10.24963/ijcai.2020/537","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2020/537","pdf_url":"https://www.ijcai.org/proceedings/2020/0537.pdf","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education","score":0.7699999809265137}],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3006188107.pdf","grobid_xml":"https://content.openalex.org/works/W3006188107.grobid-xml"},"referenced_works_count":41,"referenced_works":["https://openalex.org/W51796616","https://openalex.org/W59020575","https://openalex.org/W1279140112","https://openalex.org/W1520957705","https://openalex.org/W1536183618","https://openalex.org/W2078508369","https://openalex.org/W2111353076","https://openalex.org/W2119409989","https://openalex.org/W2132453167","https://openalex.org/W2154474435","https://openalex.org/W2228826686","https://openalex.org/W2321696644","https://openalex.org/W2413794162","https://openalex.org/W2427527485","https://openalex.org/W2577946330","https://openalex.org/W2606964149","https://openalex.org/W2608787653","https://openalex.org/W2786395785","https://openalex.org/W2884967713","https://openalex.org/W2898700502","https://openalex.org/W2903190877","https://openalex.org/W2914245851","https://openalex.org/W2947000318","https://openalex.org/W2950246755","https://openalex.org/W2952862139","https://openalex.org/W2962790689","https://openalex.org/W2963005248","https://openalex.org/W2963117013","https://openalex.org/W2963542740","https://openalex.org/W2964222271","https://openalex.org/W2965373594","https://openalex.org/W2968433472","https://openalex.org/W2969159802","https://openalex.org/W2971068072","https://openalex.org/W2971107062","https://openalex.org/W2986920492","https://openalex.org/W2995129620","https://openalex.org/W2995359496","https://openalex.org/W2997686430","https://openalex.org/W2997789497","https://openalex.org/W2998370259"],"related_works":["https://openalex.org/W3034830866","https://openalex.org/W2965373594","https://openalex.org/W2963341956","https://openalex.org/W3030163527","https://openalex.org/W2971869958","https://openalex.org/W1597904069","https://openalex.org/W2890894339","https://openalex.org/W2250539671","https://openalex.org/W2004652576","https://openalex.org/W985559363","https://openalex.org/W2187073983","https://openalex.org/W2342532162","https://openalex.org/W1590983754","https://openalex.org/W2133422947","https://openalex.org/W1498269318","https://openalex.org/W82332555","https://openalex.org/W180413045","https://openalex.org/W2079322253","https://openalex.org/W2272410705","https://openalex.org/W2034068590"],"abstract_inverted_index":{"Beginning":[0],"with":[1,15],"McCarthy's":[2],"Advice":[3],"Taker":[4],"(1959),":[5],"AI":[6],"has":[7,38],"pursued":[8],"the":[9,21,29,56,93,131],"goal":[10],"of":[11,100],"providing":[12],"a":[13,32,40,49,68,154],"system":[14,22],"explicit,":[16],"general":[17],"knowledge":[18,30],"and":[19,58,112,139,180],"having":[20],"reason":[23,75],"over":[24,79,103,166],"that":[25,88,97,130],"knowledge.":[26],"However,":[27],"expressing":[28],"in":[31,169,172,183],"formal":[33,69],"(logical":[34],"or":[35],"probabilistic)":[36],"representation":[37],"been":[39],"major":[41],"obstacle":[42],"to":[43,52,74,114,135,140],"this":[44,53,98],"research.":[45],"This":[46,171],"paper":[47],"investigates":[48],"modern":[50],"approach":[51],"problem":[54],"where":[55],"facts":[57],"rules":[59],"are":[60,149],"provided":[61],"as":[62,151,160],"natural":[63,145],"language":[64,104],"sentences,":[65],"thus":[66],"bypassing":[67],"representation.":[70],"We":[71,127],"train":[72],"transformers":[73],"(or":[76],"emulate":[77],"reasoning)":[78],"these":[80],"sentences":[81],"using":[82],"synthetically":[83],"generated":[84],"data.":[85],"Our":[86],"models,":[87],"we":[89],"call":[90],"RuleTakers,":[91],"provide":[92],"first":[94],"empirical":[95],"demonstration":[96],"kind":[99],"soft":[101],"reasoning":[102,182],"is":[105],"learnable,":[106],"can":[107],"achieve":[108],"high":[109],"(99%)":[110],"accuracy,":[111],"generalizes":[113],"test":[115],"data":[116],"requiring":[117],"substantially":[118],"deeper":[119],"chaining":[120],"than":[121],"seen":[122],"during":[123],"training":[124],"(95%+":[125],"scores).":[126],"also":[128],"demonstrate":[129],"models":[132],"transfer":[133],"well":[134],"two":[136],"hand-authored":[137],"rulebases,":[138],"rulebases":[141],"paraphrased":[142],"into":[143],"more":[144],"language.":[146,170],"These":[147],"findings":[148],"significant":[150],"it":[152],"suggests":[153,174],"new":[155,175],"role":[156],"for":[157,177],"transformers,":[158],"namely":[159],"limited":[161],"\"soft":[162],"theorem":[163],"provers\"":[164],"operating":[165],"explicit":[167],"theories":[168],"turn":[173],"possibilities":[176],"explainability,":[178],"correctability,":[179],"counterfactual":[181],"question-answering.":[184]},"counts_by_year":[{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":2},{"year":2021,"cited_by_count":5},{"year":2020,"cited_by_count":8}],"updated_date":"2026-02-09T09:26:11.010843","created_date":"2025-10-10T00:00:00"}