{"id":"https://openalex.org/W3035083896","doi":"https://doi.org/10.24963/ijcai.2020/520","title":"Towards Fully 8-bit Integer Inference for the Transformer Model","display_name":"Towards Fully 8-bit Integer Inference for the Transformer Model","publication_year":2020,"publication_date":"2020-07-01","ids":{"openalex":"https://openalex.org/W3035083896","doi":"https://doi.org/10.24963/ijcai.2020/520","mag":"3035083896"},"language":"en","primary_location":{"id":"doi:10.24963/ijcai.2020/520","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2020/520","pdf_url":"https://www.ijcai.org/proceedings/2020/0520.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":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://www.ijcai.org/proceedings/2020/0520.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100754384","display_name":"Ye Lin","orcid":"https://orcid.org/0000-0003-3247-7084"},"institutions":[{"id":"https://openalex.org/I9224756","display_name":"Northeastern University","ror":"https://ror.org/03awzbc87","country_code":"CN","type":"education","lineage":["https://openalex.org/I9224756"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ye Lin","raw_affiliation_strings":["Northeastern University, Shenyang, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Northeastern University, Shenyang, China","institution_ids":["https://openalex.org/I9224756"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101595086","display_name":"Yanyang Li","orcid":"https://orcid.org/0000-0002-6534-0618"},"institutions":[{"id":"https://openalex.org/I9224756","display_name":"Northeastern University","ror":"https://ror.org/03awzbc87","country_code":"CN","type":"education","lineage":["https://openalex.org/I9224756"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yanyang Li","raw_affiliation_strings":["Northeastern University, Shenyang, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Northeastern University, Shenyang, China","institution_ids":["https://openalex.org/I9224756"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5051358673","display_name":"Tengbo Liu","orcid":null},"institutions":[{"id":"https://openalex.org/I4210131870","display_name":"Institute of Psychology, Chinese Academy of Sciences","ror":"https://ror.org/03j7v5j15","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210131870"]},{"id":"https://openalex.org/I9224756","display_name":"Northeastern University","ror":"https://ror.org/03awzbc87","country_code":"CN","type":"education","lineage":["https://openalex.org/I9224756"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tengbo Liu","raw_affiliation_strings":["CAS Key Laboratory of Behavioral Science, Institute of Psychology, CAS, Beijing, China","Northeastern University, Shenyang, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"CAS Key Laboratory of Behavioral Science, Institute of Psychology, CAS, Beijing, China","institution_ids":["https://openalex.org/I4210131870"]},{"raw_affiliation_string":"Northeastern University, Shenyang, China","institution_ids":["https://openalex.org/I9224756"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100600701","display_name":"Tong Xiao","orcid":"https://orcid.org/0000-0002-5842-6501"},"institutions":[{"id":"https://openalex.org/I9224756","display_name":"Northeastern University","ror":"https://ror.org/03awzbc87","country_code":"CN","type":"education","lineage":["https://openalex.org/I9224756"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tong Xiao","raw_affiliation_strings":["NiuTrans Research, Shenyang, China","Northeastern University, Shenyang, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NiuTrans Research, Shenyang, China","institution_ids":[]},{"raw_affiliation_string":"Northeastern University, Shenyang, China","institution_ids":["https://openalex.org/I9224756"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012126405","display_name":"Tongran Liu","orcid":null},"institutions":[{"id":"https://openalex.org/I4210131870","display_name":"Institute of Psychology, Chinese Academy of Sciences","ror":"https://ror.org/03j7v5j15","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210131870"]},{"id":"https://openalex.org/I9224756","display_name":"Northeastern University","ror":"https://ror.org/03awzbc87","country_code":"CN","type":"education","lineage":["https://openalex.org/I9224756"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tongran Liu","raw_affiliation_strings":["CAS Key Laboratory of Behavioral Science, Institute of Psychology, CAS, Beijing, China","Northeastern University, Shenyang, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"CAS Key Laboratory of Behavioral Science, Institute of Psychology, CAS, Beijing, China","institution_ids":["https://openalex.org/I4210131870"]},{"raw_affiliation_string":"Northeastern University, Shenyang, China","institution_ids":["https://openalex.org/I9224756"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100370155","display_name":"Jingbo Zhu","orcid":"https://orcid.org/0000-0002-6537-7007"},"institutions":[{"id":"https://openalex.org/I9224756","display_name":"Northeastern University","ror":"https://ror.org/03awzbc87","country_code":"CN","type":"education","lineage":["https://openalex.org/I9224756"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jingbo Zhu","raw_affiliation_strings":["NiuTrans Research, Shenyang, China","Northeastern University, Shenyang, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NiuTrans Research, Shenyang, China","institution_ids":[]},{"raw_affiliation_string":"Northeastern University, Shenyang, China","institution_ids":["https://openalex.org/I9224756"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":29,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"3759","last_page":"3765"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998000264167786,"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"}},"topics":[{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998000264167786,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9948999881744385,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9937000274658203,"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/softmax-function","display_name":"Softmax function","score":0.7492796778678894},{"id":"https://openalex.org/keywords/floating-point","display_name":"Floating point","score":0.6464430093765259},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.6147669553756714},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6071512699127197},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5764257907867432},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5752193927764893},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.4743037223815918},{"id":"https://openalex.org/keywords/memory-footprint","display_name":"Memory footprint","score":0.4287254810333252},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4120529890060425},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.30163276195526123},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.15912103652954102},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.11577442288398743},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.11409950256347656}],"concepts":[{"id":"https://openalex.org/C188441871","wikidata":"https://www.wikidata.org/wiki/Q7554146","display_name":"Softmax function","level":3,"score":0.7492796778678894},{"id":"https://openalex.org/C84211073","wikidata":"https://www.wikidata.org/wiki/Q117879","display_name":"Floating point","level":2,"score":0.6464430093765259},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.6147669553756714},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6071512699127197},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5764257907867432},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5752193927764893},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4743037223815918},{"id":"https://openalex.org/C74912251","wikidata":"https://www.wikidata.org/wiki/Q6815727","display_name":"Memory footprint","level":2,"score":0.4287254810333252},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4120529890060425},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.30163276195526123},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.15912103652954102},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.11577442288398743},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.11409950256347656},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.24963/ijcai.2020/520","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2020/520","pdf_url":"https://www.ijcai.org/proceedings/2020/0520.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"}],"best_oa_location":{"id":"doi:10.24963/ijcai.2020/520","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2020/520","pdf_url":"https://www.ijcai.org/proceedings/2020/0520.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":[],"awards":[{"id":"https://openalex.org/G6893471174","display_name":null,"funder_award_id":"61876035","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G8440137829","display_name":"\u9762\u5411\u8d44\u6e90\u7a00\u7f3a\u578b\u8bed\u8a00\u7684\u673a\u5668\u7ffb\u8bd1\u7406\u8bba\u65b9\u6cd5\u53ca\u5173\u952e\u6280\u672f\u7814\u7a76","funder_award_id":"61732005","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3035083896.pdf","grobid_xml":"https://content.openalex.org/works/W3035083896.grobid-xml"},"referenced_works_count":29,"referenced_works":["https://openalex.org/W1677182931","https://openalex.org/W2194775991","https://openalex.org/W2604319603","https://openalex.org/W2763421725","https://openalex.org/W2777406049","https://openalex.org/W2793416812","https://openalex.org/W2797162333","https://openalex.org/W2803739089","https://openalex.org/W2899063892","https://openalex.org/W2947946877","https://openalex.org/W2948223045","https://openalex.org/W2948798935","https://openalex.org/W2955646770","https://openalex.org/W2962944188","https://openalex.org/W2963122961","https://openalex.org/W2963542740","https://openalex.org/W2963631907","https://openalex.org/W2963809228","https://openalex.org/W2965046076","https://openalex.org/W2981910001","https://openalex.org/W2997751980","https://openalex.org/W3164612223","https://openalex.org/W4288337707","https://openalex.org/W4295262505","https://openalex.org/W4385245566","https://openalex.org/W6754905691","https://openalex.org/W6795876147","https://openalex.org/W6803771590","https://openalex.org/W6864014924"],"related_works":["https://openalex.org/W3107204728","https://openalex.org/W4287591324","https://openalex.org/W3108503355","https://openalex.org/W3090555870","https://openalex.org/W4226420367","https://openalex.org/W2928062709","https://openalex.org/W4289729660","https://openalex.org/W2887023857","https://openalex.org/W2950000202","https://openalex.org/W4321472478"],"abstract_inverted_index":{"8-bit":[0,73,116],"integer":[1,74],"inference,":[2],"as":[3,104,106],"a":[4,60],"promising":[5],"direction":[6],"in":[7,38,43],"reducing":[8],"both":[9],"the":[10,24,64,89,107,114,123],"latency":[11],"and":[12,45,51,100],"storage":[13],"of":[14,49],"deep":[15],"neural":[16],"networks,":[17],"has":[18],"made":[19],"great":[20],"progress":[21],"recently.":[22],"On":[23],"other":[25],"hand,":[26],"previous":[27],"systems":[28],"still":[29],"rely":[30],"on":[31,63,95],"32-bit":[32],"floating":[33,124],"point":[34,125],"for":[35],"certain":[36],"functions":[37],"complex":[39],"models":[40],"(e.g.,":[41],"Softmax":[42],"Transformer),":[44],"make":[46],"heavy":[47],"use":[48],"quantization":[50],"de-quantization.":[52],"In":[53],"this":[54],"work,":[55],"we":[56],"show":[57,112],"that":[58,113],"after":[59],"principled":[61],"modification":[62],"Transformer":[65,117],"architecture,":[66],"dubbed":[67],"Integer":[68],"Transformer,":[69],"an":[70],"(almost)":[71],"fully":[72,115],"inference":[75],"algorithm":[76],"Scale":[77],"Propagation":[78],"could":[79],"be":[80],"derived.":[81],"De-quantization":[82],"is":[83],"adopted":[84],"when":[85],"necessary,":[86],"which":[87],"makes":[88],"network":[90],"more":[91],"efficient.":[92],"Our":[93],"experiments":[94],"WMT16":[96],"En<->Ro,":[97],"WMT14":[98],"En<->De":[99],"En->Fr":[101],"translation":[102],"tasks":[103],"well":[105],"WikiText-103":[108],"language":[109],"modelling":[110],"task":[111],"system":[118],"achieves":[119],"comparable":[120],"performance":[121],"with":[122],"baseline":[126],"but":[127],"requires":[128],"nearly":[129],"4x":[130],"less":[131],"memory":[132],"footprint.":[133]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":5},{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":8},{"year":2022,"cited_by_count":4},{"year":2021,"cited_by_count":7},{"year":2020,"cited_by_count":2}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}