{"id":"https://openalex.org/W4285601701","doi":"https://doi.org/10.24963/ijcai.2022/164","title":"FQ-ViT: Post-Training Quantization for Fully Quantized Vision Transformer","display_name":"FQ-ViT: Post-Training Quantization for Fully Quantized Vision Transformer","publication_year":2022,"publication_date":"2022-07-01","ids":{"openalex":"https://openalex.org/W4285601701","doi":"https://doi.org/10.24963/ijcai.2022/164"},"language":"en","primary_location":{"id":"doi:10.24963/ijcai.2022/164","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2022/164","pdf_url":"https://www.ijcai.org/proceedings/2022/0164.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":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://www.ijcai.org/proceedings/2022/0164.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5104094924","display_name":"Yang Lin","orcid":null},"institutions":[{"id":"https://openalex.org/I4210109870","display_name":"Vi Technology (United States)","ror":"https://ror.org/016mnbp44","country_code":"US","type":"company","lineage":["https://openalex.org/I4210109870"]},{"id":"https://openalex.org/I4401726805","display_name":"Megvii (China)","ror":"https://ror.org/040b32p69","country_code":null,"type":"company","lineage":["https://openalex.org/I4401726805"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yang Lin","raw_affiliation_strings":["MEGVII Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"MEGVII Technology","institution_ids":["https://openalex.org/I4210109870","https://openalex.org/I4401726805"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100425850","display_name":"Tianyu Zhang","orcid":"https://orcid.org/0000-0002-3908-7962"},"institutions":[{"id":"https://openalex.org/I4210109870","display_name":"Vi Technology (United States)","ror":"https://ror.org/016mnbp44","country_code":"US","type":"company","lineage":["https://openalex.org/I4210109870"]},{"id":"https://openalex.org/I4401726805","display_name":"Megvii (China)","ror":"https://ror.org/040b32p69","country_code":null,"type":"company","lineage":["https://openalex.org/I4401726805"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tianyu Zhang","raw_affiliation_strings":["MEGVII Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"MEGVII Technology","institution_ids":["https://openalex.org/I4210109870","https://openalex.org/I4401726805"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111194558","display_name":"Peiqin Sun","orcid":null},"institutions":[{"id":"https://openalex.org/I4210109870","display_name":"Vi Technology (United States)","ror":"https://ror.org/016mnbp44","country_code":"US","type":"company","lineage":["https://openalex.org/I4210109870"]},{"id":"https://openalex.org/I4401726805","display_name":"Megvii (China)","ror":"https://ror.org/040b32p69","country_code":null,"type":"company","lineage":["https://openalex.org/I4401726805"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Peiqin Sun","raw_affiliation_strings":["MEGVII Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"MEGVII Technology","institution_ids":["https://openalex.org/I4210109870","https://openalex.org/I4401726805"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100414977","display_name":"Zheng Li","orcid":"https://orcid.org/0000-0002-1447-4448"},"institutions":[{"id":"https://openalex.org/I4210109870","display_name":"Vi Technology (United States)","ror":"https://ror.org/016mnbp44","country_code":"US","type":"company","lineage":["https://openalex.org/I4210109870"]},{"id":"https://openalex.org/I4401726805","display_name":"Megvii (China)","ror":"https://ror.org/040b32p69","country_code":null,"type":"company","lineage":["https://openalex.org/I4401726805"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zheng Li","raw_affiliation_strings":["MEGVII Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"MEGVII Technology","institution_ids":["https://openalex.org/I4210109870","https://openalex.org/I4401726805"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5008993224","display_name":"Shuchang Zhou","orcid":"https://orcid.org/0000-0002-7294-5869"},"institutions":[{"id":"https://openalex.org/I4210109870","display_name":"Vi Technology (United States)","ror":"https://ror.org/016mnbp44","country_code":"US","type":"company","lineage":["https://openalex.org/I4210109870"]},{"id":"https://openalex.org/I4401726805","display_name":"Megvii (China)","ror":"https://ror.org/040b32p69","country_code":null,"type":"company","lineage":["https://openalex.org/I4401726805"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shuchang Zhou","raw_affiliation_strings":["MEGVII Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"MEGVII Technology","institution_ids":["https://openalex.org/I4210109870","https://openalex.org/I4401726805"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":12.5745,"has_fulltext":false,"cited_by_count":158,"citation_normalized_percentile":{"value":0.98901825,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":99,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"1173","last_page":"1179"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9994999766349792,"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":0.9994999766349792,"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/T12702","display_name":"Brain Tumor Detection and Classification","score":0.9975000023841858,"subfield":{"id":"https://openalex.org/subfields/2808","display_name":"Neurology"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9940000176429749,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/softmax-function","display_name":"Softmax function","score":0.703532338142395},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.6868132948875427},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.6709187030792236},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6690168976783752},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.6140832901000977},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6046870946884155},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.433645099401474},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.37737977504730225},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.34334176778793335},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.09518977999687195},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.0945720374584198}],"concepts":[{"id":"https://openalex.org/C188441871","wikidata":"https://www.wikidata.org/wiki/Q7554146","display_name":"Softmax function","level":3,"score":0.703532338142395},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.6868132948875427},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.6709187030792236},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6690168976783752},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.6140832901000977},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6046870946884155},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.433645099401474},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.37737977504730225},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.34334176778793335},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.09518977999687195},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0945720374584198},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.24963/ijcai.2022/164","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2022/164","pdf_url":"https://www.ijcai.org/proceedings/2022/0164.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"}],"best_oa_location":{"id":"doi:10.24963/ijcai.2022/164","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2022/164","pdf_url":"https://www.ijcai.org/proceedings/2022/0164.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/W4285601701.pdf","grobid_xml":"https://content.openalex.org/works/W4285601701.grobid-xml"},"referenced_works_count":26,"referenced_works":["https://openalex.org/W1836465849","https://openalex.org/W1861492603","https://openalex.org/W2163605009","https://openalex.org/W2194775991","https://openalex.org/W2469490737","https://openalex.org/W2903972532","https://openalex.org/W2946355854","https://openalex.org/W2962298324","https://openalex.org/W2963122961","https://openalex.org/W2998183051","https://openalex.org/W2998218113","https://openalex.org/W3035078980","https://openalex.org/W3094502228","https://openalex.org/W3096609285","https://openalex.org/W3105966348","https://openalex.org/W3138516171","https://openalex.org/W3168124404","https://openalex.org/W3170233084","https://openalex.org/W3170841864","https://openalex.org/W3170874841","https://openalex.org/W3211787299","https://openalex.org/W4214588794","https://openalex.org/W4287812978","https://openalex.org/W4297813615","https://openalex.org/W4308536459","https://openalex.org/W4394666973"],"related_works":["https://openalex.org/W2977314777","https://openalex.org/W2743258233","https://openalex.org/W4307834408","https://openalex.org/W4320925816","https://openalex.org/W3208883981","https://openalex.org/W2758063741","https://openalex.org/W2908861653","https://openalex.org/W2613736958","https://openalex.org/W2997969508","https://openalex.org/W2767651786"],"abstract_inverted_index":{"Network":[0],"quantization":[1,18,103],"significantly":[2],"reduces":[3],"model":[4],"inference":[5,73,99],"complexity":[6,74],"and":[7,29,59,72,97,104,114,146],"has":[8],"been":[9,21],"widely":[10],"used":[11],"in":[12,56,87],"real-world":[13],"deployments.":[14],"However,":[15],"most":[16],"existing":[17],"methods":[19],"have":[20],"developed":[22],"mainly":[23],"on":[24,110,132,144,154,169],"Convolutional":[25],"Neural":[26],"Networks":[27],"(CNNs),":[28],"suffer":[30],"severe":[31],"degradation":[32,71,167],"when":[33],"applied":[34],"to":[35,67,94,163],"fully":[36,76,170],"quantized":[37,77,171],"vision":[38,78,172],"transformers.":[39,79,173],"In":[40,80],"this":[41],"work,":[42],"we":[43,90,137,159],"demonstrate":[44],"that":[45,96,117],"many":[46],"of":[47,52,75],"these":[48],"difficulties":[49],"arise":[50],"because":[51],"serious":[53],"inter-channel":[54],"variation":[55],"LayerNorm":[57],"inputs,":[58],"present,":[60],"Power-of-Two":[61],"Factor":[62],"(PTF),":[63],"a":[64],"systematic":[65],"method":[66],"reduce":[68],"the":[69,105,161],"performance":[70],"addition,":[81],"observing":[82],"an":[83],"extreme":[84],"non-uniform":[85],"distribution":[86],"attention":[88,133],"maps,":[89],"propose":[91],"Log-Int-Softmax":[92],"(LIS)":[93],"sustain":[95],"simplify":[98],"by":[100],"using":[101,129],"4-bit":[102],"BitShift":[106],"operator.":[107],"Comprehensive":[108],"experiments":[109],"various":[111],"transformer-based":[112],"architectures":[113],"benchmarks":[115],"show":[116],"our":[118,157],"Fully":[119],"Quantized":[120],"Vision":[121],"Transformer":[122],"(FQ-ViT)":[123],"outperforms":[124],"previous":[125],"works":[126],"while":[127],"even":[128],"lower":[130],"bit-width":[131],"maps.":[134],"For":[135],"instance,":[136],"reach":[138],"84.89%":[139],"top-1":[140],"accuracy":[141,166],"with":[142,149],"ViT-L":[143],"ImageNet":[145],"50.8":[147],"mAP":[148],"Cascade":[150],"Mask":[151],"R-CNN":[152],"(Swin-S)":[153],"COCO.":[155],"To":[156],"knowledge,":[158],"are":[160],"first":[162],"achieve":[164],"lossless":[165],"(~1%)":[168],"The":[174],"code":[175],"is":[176],"available":[177],"at":[178],"https://github.com/megvii-research/FQ-ViT.":[179]},"counts_by_year":[{"year":2026,"cited_by_count":22},{"year":2025,"cited_by_count":70},{"year":2024,"cited_by_count":41},{"year":2023,"cited_by_count":25}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}