{"id":"https://openalex.org/W4206235551","doi":"https://doi.org/10.1080/09540091.2021.2024510","title":"Multi-distribution noise quantisation: an extreme compression scheme for transformer according to parameter distribution","display_name":"Multi-distribution noise quantisation: an extreme compression scheme for transformer according to parameter distribution","publication_year":2022,"publication_date":"2022-01-17","ids":{"openalex":"https://openalex.org/W4206235551","doi":"https://doi.org/10.1080/09540091.2021.2024510"},"language":"en","primary_location":{"id":"doi:10.1080/09540091.2021.2024510","is_oa":true,"landing_page_url":"https://doi.org/10.1080/09540091.2021.2024510","pdf_url":null,"source":{"id":"https://openalex.org/S4210188800","display_name":"Connection Science","issn_l":"0954-0091","issn":["0954-0091","1360-0494"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310320547","host_organization_name":"Taylor & Francis","host_organization_lineage":["https://openalex.org/P4310320547"],"host_organization_lineage_names":["Taylor & Francis"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Connection Science","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","doaj"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1080/09540091.2021.2024510","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5085760057","display_name":"Zaiyang Yu","orcid":"https://orcid.org/0000-0002-3425-1153"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"funder","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210149211","display_name":"Institute of Semiconductors","ror":"https://ror.org/048dd0611","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210149211"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zaiyang Yu","raw_affiliation_strings":["Cognitive Computing Technology Joint Laboratory, Wave Group, Beijing, People's Republic of China","Institute of Semiconductors, Chinese Academy of Sciences, Beijing, People's Republic of China"],"affiliations":[{"raw_affiliation_string":"Cognitive Computing Technology Joint Laboratory, Wave Group, Beijing, People's Republic of China","institution_ids":[]},{"raw_affiliation_string":"Institute of Semiconductors, Chinese Academy of Sciences, Beijing, People's Republic of China","institution_ids":["https://openalex.org/I4210149211","https://openalex.org/I19820366"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100767125","display_name":"Shuang Li","orcid":"https://orcid.org/0000-0002-3089-7221"},"institutions":[{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Shuang Li","raw_affiliation_strings":["Beijing Key Laboratory of Semiconductor Neural Network Intelligent Sensing and Computing Technology, Beijing, People's Republic of China","Center of Materials Science and Optoelectronics Engineering & School of Microelectronics, University of Chinese Academy of Sciences, Beijing, People's Republic of China"],"affiliations":[{"raw_affiliation_string":"Beijing Key Laboratory of Semiconductor Neural Network Intelligent Sensing and Computing Technology, Beijing, People's Republic of China","institution_ids":[]},{"raw_affiliation_string":"Center of Materials Science and Optoelectronics Engineering & School of Microelectronics, University of Chinese Academy of Sciences, Beijing, People's Republic of China","institution_ids":["https://openalex.org/I4210165038"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5068285118","display_name":"Linjun Sun","orcid":"https://orcid.org/0000-0002-9287-9467"},"institutions":[{"id":"https://openalex.org/I4210149211","display_name":"Institute of Semiconductors","ror":"https://ror.org/048dd0611","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210149211"]},{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"funder","lineage":["https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Linjun Sun","raw_affiliation_strings":["Cognitive Computing Technology Joint Laboratory, Wave Group, Beijing, People's Republic of China","Institute of Semiconductors, Chinese Academy of Sciences, Beijing, People's Republic of China"],"affiliations":[{"raw_affiliation_string":"Cognitive Computing Technology Joint Laboratory, Wave Group, Beijing, People's Republic of China","institution_ids":[]},{"raw_affiliation_string":"Institute of Semiconductors, Chinese Academy of Sciences, Beijing, People's Republic of China","institution_ids":["https://openalex.org/I4210149211","https://openalex.org/I19820366"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100322342","display_name":"Liang Liu","orcid":"https://orcid.org/0000-0001-7910-810X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liang Liu","raw_affiliation_strings":["Cognitive Computing Technology Joint Laboratory, Wave Group, Beijing, People's Republic of China"],"affiliations":[{"raw_affiliation_string":"Cognitive Computing Technology Joint Laboratory, Wave Group, Beijing, People's Republic of China","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100664238","display_name":"Haining Wang","orcid":"https://orcid.org/0000-0002-5843-793X"},"institutions":[{"id":"https://openalex.org/I37448385","display_name":"China People's Public Security University","ror":"https://ror.org/05twya590","country_code":"CN","type":"education","lineage":["https://openalex.org/I37448385"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wang Haining","raw_affiliation_strings":["School of Police Administration, People's Public Security University of China, Beijing, People's Republic of China"],"affiliations":[{"raw_affiliation_string":"School of Police Administration, People's Public Security University of China, Beijing, People's Republic of China","institution_ids":["https://openalex.org/I37448385"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5068285118","https://openalex.org/A5100767125"],"corresponding_institution_ids":["https://openalex.org/I19820366","https://openalex.org/I4210149211","https://openalex.org/I4210165038"],"apc_list":{"value":1270,"currency":"USD","value_usd":1270},"apc_paid":{"value":1270,"currency":"USD","value_usd":1270},"fwci":1.4092,"has_fulltext":false,"cited_by_count":14,"citation_normalized_percentile":{"value":0.8196761,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":99},"biblio":{"volume":"34","issue":"1","first_page":"990","last_page":"1004"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9940999746322632,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9940999746322632,"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/T10028","display_name":"Topic Modeling","score":0.9929999709129333,"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":0.9929999709129333,"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":0.7394404411315918},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.5840149521827698},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.544325053691864},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5119246244430542},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.5031127333641052},{"id":"https://openalex.org/keywords/piecewise","display_name":"Piecewise","score":0.4834795594215393},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.4176441729068756},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.33561980724334717},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.11640971899032593}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7394404411315918},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5840149521827698},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.544325053691864},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5119246244430542},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.5031127333641052},{"id":"https://openalex.org/C164660894","wikidata":"https://www.wikidata.org/wiki/Q2037833","display_name":"Piecewise","level":2,"score":0.4834795594215393},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.4176441729068756},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.33561980724334717},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.11640971899032593},{"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/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1080/09540091.2021.2024510","is_oa":true,"landing_page_url":"https://doi.org/10.1080/09540091.2021.2024510","pdf_url":null,"source":{"id":"https://openalex.org/S4210188800","display_name":"Connection Science","issn_l":"0954-0091","issn":["0954-0091","1360-0494"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310320547","host_organization_name":"Taylor & Francis","host_organization_lineage":["https://openalex.org/P4310320547"],"host_organization_lineage_names":["Taylor & Francis"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Connection Science","raw_type":"journal-article"},{"id":"pmh:oai:doaj.org/article:7df893e9bff044d78bee20d8f2e2af60","is_oa":true,"landing_page_url":"https://doaj.org/article/7df893e9bff044d78bee20d8f2e2af60","pdf_url":null,"source":{"id":"https://openalex.org/S112646816","display_name":"SHILAP Revista de lepidopterolog\u00eda","issn_l":"0300-5267","issn":["0300-5267","2340-4078"],"is_oa":true,"is_in_doaj":true,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Connection Science, Vol 34, Iss 1, Pp 990-1004 (2022)","raw_type":"article"}],"best_oa_location":{"id":"doi:10.1080/09540091.2021.2024510","is_oa":true,"landing_page_url":"https://doi.org/10.1080/09540091.2021.2024510","pdf_url":null,"source":{"id":"https://openalex.org/S4210188800","display_name":"Connection Science","issn_l":"0954-0091","issn":["0954-0091","1360-0494"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310320547","host_organization_name":"Taylor & Francis","host_organization_lineage":["https://openalex.org/P4310320547"],"host_organization_lineage_names":["Taylor & Francis"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Connection Science","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1427365533","display_name":null,"funder_award_id":"61901436","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":49,"referenced_works":["https://openalex.org/W1535111405","https://openalex.org/W1686810756","https://openalex.org/W2077815765","https://openalex.org/W2093647425","https://openalex.org/W2124509324","https://openalex.org/W2147646381","https://openalex.org/W2148554573","https://openalex.org/W2567289819","https://openalex.org/W2599837009","https://openalex.org/W2884150179","https://openalex.org/W2897049076","https://openalex.org/W2905559537","https://openalex.org/W2914294010","https://openalex.org/W2933138175","https://openalex.org/W2936652946","https://openalex.org/W2946232455","https://openalex.org/W2947946877","https://openalex.org/W2960643108","https://openalex.org/W2963122961","https://openalex.org/W2972555443","https://openalex.org/W2979691890","https://openalex.org/W2981751377","https://openalex.org/W2983157715","https://openalex.org/W2988841832","https://openalex.org/W3000437185","https://openalex.org/W3003175769","https://openalex.org/W3015720739","https://openalex.org/W3041838758","https://openalex.org/W3049526112","https://openalex.org/W3084016763","https://openalex.org/W3093480222","https://openalex.org/W3098576111","https://openalex.org/W3098873988","https://openalex.org/W3101493110","https://openalex.org/W3111255625","https://openalex.org/W3119189390","https://openalex.org/W3128633047","https://openalex.org/W3150864706","https://openalex.org/W3154998900","https://openalex.org/W3174422331","https://openalex.org/W3179877418","https://openalex.org/W3206736890","https://openalex.org/W4302978247","https://openalex.org/W6739901393","https://openalex.org/W6752675014","https://openalex.org/W6768091732","https://openalex.org/W6793979934","https://openalex.org/W6796487566","https://openalex.org/W6797602710"],"related_works":["https://openalex.org/W2046435967","https://openalex.org/W2385263368","https://openalex.org/W4231775656","https://openalex.org/W2347422947","https://openalex.org/W2061292372","https://openalex.org/W328874995","https://openalex.org/W2353392568","https://openalex.org/W2383646825","https://openalex.org/W2377341410","https://openalex.org/W1454600515"],"abstract_inverted_index":{"With":[0],"the":[1,15,64,67,80,90,95,107,116,126,130,134,145,169,175,189],"development":[2],"of":[3,24,58,69,97,103,129,147,171],"deep":[4],"learning,":[5],"neural":[6],"networks":[7],"are":[8],"widely":[9],"used":[10],"in":[11,100],"various":[12],"fields,":[13],"and":[14,26,49,93,123,177,192],"improved":[16],"model":[17,44,140,195],"performance":[18],"also":[19],"introduces":[20],"a":[21,31,55,85,148],"considerable":[22,56],"number":[23],"parameters":[25,70],"computations.":[27],"Model":[28],"quantisation":[29,75,98,156],"is":[30,71,164],"technique":[32],"that":[33,113,184],"turns":[34],"floating-point":[35],"computing":[36],"into":[37],"low-specific-point":[38],"computing,":[39],"which":[40],"can":[41],"effectively":[42],"reduce":[43,125],"computation":[45],"strength,":[46],"parameter":[47,91],"size,":[48],"memory":[50],"consumption":[51],"but":[52],"often":[53],"bring":[54],"loss":[57],"accuracy.":[59],"This":[60],"paper":[61],"mainly":[62],"addresses":[63],"problem":[65],"where":[66],"distribution":[68],"too":[72],"concentrated":[73],"during":[74],"aware":[76],"training":[77],"(QAT).":[78],"In":[79],"QAT":[81],"process,":[82],"we":[83,119],"use":[84],"piecewise":[86],"function":[87],"to":[88],"statistics":[89],"distributions":[92],"simulate":[94],"effect":[96],"noise":[99],"each":[101],"round":[102],"training,":[104],"based":[105],"on":[106,158,174],"statistical":[108],"results.":[109],"Experimental":[110],"results":[111],"show":[112,183],"by":[114],"quantising":[115],"Transformer":[117],"network,":[118,138],"lose":[120],"less":[121],"precision":[122,136],"significantly":[124],"storage":[127,150,190],"cost":[128,191],"model;":[131],"compared":[132,153],"with":[133,154],"full":[135],"LSTM":[137],"our":[139,162,172,185],"has":[141],"higher":[142],"accuracy":[143],"under":[144],"condition":[146],"similar":[149],"cost.":[151],"Meanwhile,":[152],"other":[155],"methods":[157],"language":[159],"modelling":[160],"task,":[161],"approach":[163],"more":[165],"accurate.":[166],"We":[167],"validated":[168],"effectiveness":[170],"policy":[173],"WikiText-103":[176],"PENN":[178],"Treebank":[179],"datasets.":[180],"The":[181],"experiments":[182],"method":[186],"extremely":[187],"compresses":[188],"maintains":[193],"high":[194],"performance.":[196]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2023,"cited_by_count":5},{"year":2022,"cited_by_count":8}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}