{"id":"https://openalex.org/W4404133542","doi":"https://doi.org/10.1145/3649329.3657314","title":"Genetic Quantization-Aware Approximation for Non-Linear Operations in Transformers","display_name":"Genetic Quantization-Aware Approximation for Non-Linear Operations in Transformers","publication_year":2024,"publication_date":"2024-06-23","ids":{"openalex":"https://openalex.org/W4404133542","doi":"https://doi.org/10.1145/3649329.3657314"},"language":"en","primary_location":{"id":"doi:10.1145/3649329.3657314","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3649329.3657314","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 61st ACM/IEEE Design Automation Conference","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5043072381","display_name":"Pingcheng Dong","orcid":"https://orcid.org/0000-0003-3975-286X"},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Pingcheng Dong","raw_affiliation_strings":["The Hong Kong University of Science and Technology, Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0003-3975-286X","affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology, Hong Kong, Hong Kong, China","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5056333723","display_name":"Yonghao Tan","orcid":"https://orcid.org/0000-0001-5372-5863"},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Yonghao Tan","raw_affiliation_strings":["The Hong Kong University of Science and Technology, Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0001-5372-5863","affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology, Hong Kong, Hong Kong, China","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101818657","display_name":"Dong Zhang","orcid":"https://orcid.org/0000-0002-4543-2179"},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Dong Zhang","raw_affiliation_strings":["The Hong Kong University of Science and Technology, Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0002-4543-2179","affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology, Hong Kong, Hong Kong, China","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5037238501","display_name":"Tianwei Ni","orcid":"https://orcid.org/0009-0005-1045-6698"},"institutions":[{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tianwei Ni","raw_affiliation_strings":["Zhejiang University, Hangzhou, Zhejiang, China"],"raw_orcid":"https://orcid.org/0009-0005-1045-6698","affiliations":[{"raw_affiliation_string":"Zhejiang University, Hangzhou, Zhejiang, China","institution_ids":["https://openalex.org/I76130692"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5086868241","display_name":"Xuejiao Liu","orcid":"https://orcid.org/0009-0007-6788-1593"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xuejiao Liu","raw_affiliation_strings":["AI Chip Center for Emerging Smart System (ACCESS), Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0009-0007-6788-1593","affiliations":[{"raw_affiliation_string":"AI Chip Center for Emerging Smart System (ACCESS), Hong Kong, Hong Kong, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100445929","display_name":"Yu Liu","orcid":"https://orcid.org/0000-0002-1170-3858"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu Liu","raw_affiliation_strings":["AI Chip Center for Emerging Smart System (ACCESS), Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0002-1170-3858","affiliations":[{"raw_affiliation_string":"AI Chip Center for Emerging Smart System (ACCESS), Hong Kong, Hong Kong, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015735396","display_name":"Peng Luo","orcid":"https://orcid.org/0009-0005-0256-9545"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Peng Luo","raw_affiliation_strings":["AI Chip Center for Emerging Smart System (ACCESS), Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0009-0005-0256-9545","affiliations":[{"raw_affiliation_string":"AI Chip Center for Emerging Smart System (ACCESS), Hong Kong, Hong Kong, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5052554057","display_name":"Luhong Liang","orcid":"https://orcid.org/0009-0001-4190-006X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Luhong Liang","raw_affiliation_strings":["AI Chip Center for Emerging Smart System (ACCESS), Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0009-0001-4190-006X","affiliations":[{"raw_affiliation_string":"AI Chip Center for Emerging Smart System (ACCESS), Hong Kong, Hong Kong, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016722419","display_name":"Shih-Yang Liu","orcid":"https://orcid.org/0000-0003-1997-0843"},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Shih-Yang Liu","raw_affiliation_strings":["The Hong Kong University of Science and Technology, Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0003-1997-0843","affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology, Hong Kong, Hong Kong, China","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5049883737","display_name":"Xijie Huang","orcid":"https://orcid.org/0000-0001-5508-5433"},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Xijie Huang","raw_affiliation_strings":["The Hong Kong University of Science and Technology, Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0001-5508-5433","affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology, Hong Kong, Hong Kong, China","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5076801334","display_name":"Huaiyu Zhu","orcid":"https://orcid.org/0000-0001-6918-4088"},"institutions":[{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Huaiyu Zhu","raw_affiliation_strings":["Zhejiang University, Hangzhou, Zhejiang, China"],"raw_orcid":"https://orcid.org/0000-0001-6918-4088","affiliations":[{"raw_affiliation_string":"Zhejiang University, Hangzhou, Zhejiang, China","institution_ids":["https://openalex.org/I76130692"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5081883922","display_name":"Yun Pan","orcid":"https://orcid.org/0000-0002-9335-4291"},"institutions":[{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yun Pan","raw_affiliation_strings":["Zhejiang University, Hangzhou, Zhejiang, China"],"raw_orcid":"https://orcid.org/0000-0002-9335-4291","affiliations":[{"raw_affiliation_string":"Zhejiang University, Hangzhou, Zhejiang, China","institution_ids":["https://openalex.org/I76130692"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5069123107","display_name":"Fengwei An","orcid":"https://orcid.org/0000-0002-7554-7938"},"institutions":[{"id":"https://openalex.org/I3045169105","display_name":"Southern University of Science and Technology","ror":"https://ror.org/049tv2d57","country_code":"CN","type":"education","lineage":["https://openalex.org/I3045169105"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Fengwei An","raw_affiliation_strings":["Southern University of Science and Technology, Shenzhen, Guangdong, China"],"raw_orcid":"https://orcid.org/0000-0002-7554-7938","affiliations":[{"raw_affiliation_string":"Southern University of Science and Technology, Shenzhen, Guangdong, China","institution_ids":["https://openalex.org/I3045169105"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5077687075","display_name":"Kwang\u2010Ting Cheng","orcid":"https://orcid.org/0000-0002-3885-4912"},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Kwang-Ting Cheng","raw_affiliation_strings":["The Hong Kong University of Science and Technology, Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0002-3885-4912","affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology, Hong Kong, Hong Kong, China","institution_ids":["https://openalex.org/I200769079"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":11,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9973999857902527,"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/T10320","display_name":"Neural Networks and Applications","score":0.9973999857902527,"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/T11975","display_name":"Evolutionary Algorithms and Applications","score":0.989300012588501,"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/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9872999787330627,"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/quantization","display_name":"Quantization (signal processing)","score":0.6313962936401367},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5727787613868713},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5003812313079834},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.2573341727256775},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.1948796510696411},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.1638517677783966},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.06944212317466736}],"concepts":[{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.6313962936401367},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5727787613868713},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5003812313079834},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.2573341727256775},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.1948796510696411},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.1638517677783966},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.06944212317466736}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3649329.3657314","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3649329.3657314","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 61st ACM/IEEE Design Automation Conference","raw_type":"proceedings-article"},{"id":"pmh:oai:repository.hkust.edu.hk:1783.1-147433","is_oa":false,"landing_page_url":"http://repository.hkust.edu.hk/ir/Record/1783.1-147433","pdf_url":null,"source":{"id":"https://openalex.org/S4306401796","display_name":"Rare & Special e-Zone (The Hong Kong University of Science and Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I200769079","host_organization_name":"Hong Kong University of Science and Technology","host_organization_lineage":["https://openalex.org/I200769079"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Conference paper"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":13,"referenced_works":["https://openalex.org/W1659842140","https://openalex.org/W2593831809","https://openalex.org/W2747329762","https://openalex.org/W2913668833","https://openalex.org/W2914911817","https://openalex.org/W3170544306","https://openalex.org/W3211490618","https://openalex.org/W4205540825","https://openalex.org/W4234552385","https://openalex.org/W4320492418","https://openalex.org/W4327522612","https://openalex.org/W4386108348","https://openalex.org/W4386767094"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"Non-linear":[0],"functions":[1],"are":[2],"prevalent":[3],"in":[4,31],"Transformers":[5],"and":[6,12,27,47,89,109],"their":[7],"lightweight":[8],"variants,":[9],"incurring":[10],"substantial":[11],"frequently":[13],"underestimated":[14],"hardware":[15],"costs.":[16],"Previous":[17],"state-of-the-art":[18],"works":[19],"optimize":[20],"these":[21],"operations":[22],"by":[23],"piece-wise":[24],"linear":[25,90],"approximation":[26],"store":[28],"the":[29,67,81,97,117],"parameters":[30,68],"look-up":[32],"tables":[33],"(LUT),":[34],"but":[35],"most":[36],"of":[37,50,99,107,113],"them":[38],"require":[39],"unfriendly":[40],"high-precision":[41,118],"arithmetics":[42],"such":[43],"as":[44],"FP/INT":[45,119],"32":[46,120],"lack":[48],"consideration":[49],"integer-only":[51],"INT":[52],"quantization.":[53],"This":[54],"paper":[55],"proposed":[56,94],"a":[57,110],"genetic":[58],"LUT-Approximation":[59,101],"algorithm":[60],"namely":[61],"GQA-LUT":[62,76,95],"that":[63,75,102],"can":[64],"automatically":[65],"determine":[66],"with":[69],"quantization":[70],"awareness.":[71],"The":[72],"results":[73],"demonstrate":[74],"achieves":[77,103],"negligible":[78],"degradation":[79],"on":[80],"challenging":[82],"semantic":[83],"segmentation":[84],"task":[85],"for":[86],"both":[87],"vanilla":[88],"Transformer":[91],"models.":[92],"Besides,":[93],"enables":[96],"employment":[98],"INT8-based":[100],"an":[104],"area":[105],"savings":[106],"81.3~81.7%":[108],"power":[111],"reduction":[112],"79.3~80.2%":[114],"compared":[115],"to":[116],"alternatives.":[121],"Code":[122],"is":[123],"available":[124],"at":[125],"https://github.com/PingchengDong/GQA-LUT.":[126]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":8},{"year":2024,"cited_by_count":2}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}
