{"id":"https://openalex.org/W3217002002","doi":null,"title":"SimpleTron: Eliminating Softmax from Attention Computation.","display_name":"SimpleTron: Eliminating Softmax from Attention Computation.","publication_year":2021,"publication_date":"2021-12-02","ids":{"openalex":"https://openalex.org/W3217002002","mag":"3217002002"},"language":"en","primary_location":{"id":"mag:3217002002","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/pdf/2111.15588.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},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":[],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"http://arxiv.org/pdf/2111.15588.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5014516814","display_name":"Uladzislau Yorsh","orcid":"https://orcid.org/0009-0003-2361-8073","observed_orcids":["https://orcid.org/0009-0003-2361-8073"]},"institutions":[{"id":"https://openalex.org/I44504214","display_name":"Czech Technical University in Prague","ror":"https://ror.org/03kqpb082","country_code":"CZ","type":"education","lineage":["https://openalex.org/I44504214"]}],"countries":["CZ"],"is_corresponding":false,"raw_author_name":"Uladzislau Yorsh","raw_affiliation_strings":["Czech Technical Univ. in Prague"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Czech Technical Univ. in Prague","institution_ids":["https://openalex.org/I44504214"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5035026045","display_name":"Pavel Kord\u00edk","orcid":"https://orcid.org/0000-0003-1433-0089","observed_orcids":["https://orcid.org/0000-0003-1433-0089"]},"institutions":[{"id":"https://openalex.org/I44504214","display_name":"Czech Technical University in Prague","ror":"https://ror.org/03kqpb082","country_code":"CZ","type":"education","lineage":["https://openalex.org/I44504214"]}],"countries":["CZ"],"is_corresponding":false,"raw_author_name":"Pavel Kord\u00edk","raw_affiliation_strings":["Czech Technical Univ. in Prague"],"raw_orcid":"https://orcid.org/0000-0003-1433-0089","affiliations":[{"raw_affiliation_string":"Czech Technical Univ. in Prague","institution_ids":["https://openalex.org/I44504214"]}]},{"author_position":"last","author":{"id":null,"display_name":"Alexander Kovalenko","orcid":"https://orcid.org/0000-0002-7194-1874","observed_orcids":["https://orcid.org/0000-0002-7194-1874"]},"institutions":[{"id":"https://openalex.org/I44504214","display_name":"Czech Technical University in Prague","ror":"https://ror.org/03kqpb082","country_code":"CZ","type":"education","lineage":["https://openalex.org/I44504214"]}],"countries":["CZ"],"is_corresponding":false,"raw_author_name":"Alexander Kovalenko","raw_affiliation_strings":["Czech Technical Univ. in Prague"],"raw_orcid":"https://orcid.org/0000-0002-7194-1874","affiliations":[{"raw_affiliation_string":"Czech Technical Univ. in Prague","institution_ids":["https://openalex.org/I44504214"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I44504214"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.801800012588501,"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":0.801800012588501,"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":0.08410000056028366,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.047200001776218414,"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/attention-mechanism","display_name":"attention mechanism","score":0.5759999752044678},{"id":"https://openalex.org/keywords/dot-product-attention","display_name":"dot-product attention","score":0.765999972820282},{"id":"https://openalex.org/keywords/transformer","display_name":"transformer models","score":0.6890000104904175},{"id":"https://openalex.org/keywords/pairwise-matching","display_name":"pairwise matching","score":0.5139999985694885}],"concepts":[{"id":"https://openalex.org/C188441871","wikidata":"https://www.wikidata.org/wiki/Q7554146","display_name":"Softmax function","level":3,"score":0.9053331017494202},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.7073513865470886},{"id":"https://openalex.org/C184898388","wikidata":"https://www.wikidata.org/wiki/Q1435712","display_name":"Pairwise comparison","level":2,"score":0.6940054297447205},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6861492991447449},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.6734287142753601},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.6583355665206909},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.510459303855896},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.45217660069465637},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.45148521661758423},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.44057998061180115},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.4168757200241089},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3423868417739868},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.30028533935546875},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.1584027111530304},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.12147089838981628},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.09646552801132202},{"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/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"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/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0},{"id":"https://openalex.org/C111472728","wikidata":"https://www.wikidata.org/wiki/Q9471","display_name":"Epistemology","level":1,"score":0.0},{"id":"https://openalex.org/C13280743","wikidata":"https://www.wikidata.org/wiki/Q131089","display_name":"Geodesy","level":1,"score":0.0},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.0},{"id":"https://openalex.org/C146978453","wikidata":"https://www.wikidata.org/wiki/Q3798668","display_name":"Aerospace engineering","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"mag:3217002002","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/pdf/2111.15588.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}],"best_oa_location":{"id":"mag:3217002002","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/pdf/2111.15588.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},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":27,"referenced_works":["https://openalex.org/W2252143850","https://openalex.org/W2626778328","https://openalex.org/W2896457183","https://openalex.org/W2937843571","https://openalex.org/W2940744433","https://openalex.org/W2986922898","https://openalex.org/W2997517014","https://openalex.org/W2999008758","https://openalex.org/W3000514857","https://openalex.org/W3007746272","https://openalex.org/W3015468748","https://openalex.org/W3030163527","https://openalex.org/W3085139254","https://openalex.org/W3091998909","https://openalex.org/W3097777922","https://openalex.org/W3103682594","https://openalex.org/W3119786062","https://openalex.org/W3120633509","https://openalex.org/W3125056032","https://openalex.org/W3134226059","https://openalex.org/W3157506437","https://openalex.org/W3162090017","https://openalex.org/W3162665866","https://openalex.org/W3173365702","https://openalex.org/W3199892937","https://openalex.org/W3207906093","https://openalex.org/W3212637808"],"related_works":["https://openalex.org/W4311034014","https://openalex.org/W3214985070","https://openalex.org/W7133231475","https://openalex.org/W3215613613","https://openalex.org/W4414581610","https://openalex.org/W3168244670","https://openalex.org/W4283204127","https://openalex.org/W4225525539","https://openalex.org/W7166831317","https://openalex.org/W4394708787"],"abstract_inverted_index":{"In":[0],"this":[1],"paper,":[2],"we":[3,50],"propose":[4],"that":[5],"the":[6,23,47,62,75],"dot":[7],"product":[8],"pairwise":[9],"matching":[10],"attention":[11,69],"layer,":[12],"which":[13],"is":[14,20],"widely":[15],"used":[16],"in":[17,27,46],"transformer-based":[18],"models,":[19],"redundant":[21],"for":[22],"model":[24],"performance.":[25],"Attention":[26],"its":[28],"original":[29],"formulation":[30],"has":[31],"to":[32,40,61],"be":[33],"rather":[34],"seen":[35],"as":[36],"a":[37,52],"human-level":[38],"tool":[39],"explore":[41],"and/or":[42],"visualize":[43],"relevancy":[44],"scores":[45],"sequences.":[48],"Instead,":[49],"present":[51],"simple":[53],"and":[54],"fast":[55],"alternative":[56],"without":[57],"any":[58],"approximation":[59],"that,":[60],"best":[63],"of":[64],"our":[65],"knowledge,":[66],"outperforms":[67],"existing":[68],"approximations":[70],"on":[71],"several":[72],"tasks":[73],"from":[74],"Long-Range":[76],"Arena":[77],"benchmark.":[78]},"counts_by_year":[],"updated_date":"2026-11-04T10:54:05.112534","created_date":"2021-12-06T00:00:00"}