{"id":"https://openalex.org/W4415142253","doi":"https://doi.org/10.1145/3786582.3786800","title":"A Selective Quantization Tuner for ONNX Models","display_name":"A Selective Quantization Tuner for ONNX Models","publication_year":2026,"publication_date":"2026-04-12","ids":{"openalex":"https://openalex.org/W4415142253","doi":"https://doi.org/10.1145/3786582.3786800"},"language":"en","primary_location":{"id":"doi:10.1145/3786582.3786800","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3786582.3786800","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the IEEE/ACM 48th International Conference on Software Engineering","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3786582.3786800","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5004785708","display_name":"Nikolaos Louloudakis","orcid":null},"institutions":[{"id":"https://openalex.org/I98677209","display_name":"University of Edinburgh","ror":"https://ror.org/01nrxwf90","country_code":"GB","type":"education","lineage":["https://openalex.org/I98677209"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Nikolaos Louloudakis","raw_affiliation_strings":["University of Edinburgh, Edinburgh, Midlothian, United Kingdom"],"raw_orcid":"https://orcid.org/0000-0002-1878-7679","affiliations":[{"raw_affiliation_string":"University of Edinburgh, Edinburgh, Midlothian, United Kingdom","institution_ids":["https://openalex.org/I98677209"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5079075574","display_name":"Ajitha Rajan","orcid":"https://orcid.org/0000-0003-3765-3075"},"institutions":[{"id":"https://openalex.org/I98677209","display_name":"University of Edinburgh","ror":"https://ror.org/01nrxwf90","country_code":"GB","type":"education","lineage":["https://openalex.org/I98677209"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Ajitha Rajan","raw_affiliation_strings":["University of Edinburgh, Edinburgh, Midlothian, United Kingdom"],"raw_orcid":"https://orcid.org/0000-0003-3765-3075","affiliations":[{"raw_affiliation_string":"University of Edinburgh, Edinburgh, Midlothian, United Kingdom","institution_ids":["https://openalex.org/I98677209"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I98677209"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.00486936,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"6","last_page":"10"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10791","display_name":"Advanced Control Systems Optimization","score":0.398499995470047,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10791","display_name":"Advanced Control Systems Optimization","score":0.398499995470047,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems 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":0.7896000146865845},{"id":"https://openalex.org/keywords/workaround","display_name":"Workaround","score":0.605400025844574},{"id":"https://openalex.org/keywords/minification","display_name":"Minification","score":0.5774000287055969},{"id":"https://openalex.org/keywords/flops","display_name":"FLOPS","score":0.4359999895095825},{"id":"https://openalex.org/keywords/pareto-optimal","display_name":"Pareto optimal","score":0.41670000553131104},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.37369999289512634},{"id":"https://openalex.org/keywords/profiling","display_name":"Profiling (computer programming)","score":0.36340001225471497},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.36090001463890076}],"concepts":[{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.7896000146865845},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.722000002861023},{"id":"https://openalex.org/C194541083","wikidata":"https://www.wikidata.org/wiki/Q457174","display_name":"Workaround","level":2,"score":0.605400025844574},{"id":"https://openalex.org/C147764199","wikidata":"https://www.wikidata.org/wiki/Q6865248","display_name":"Minification","level":2,"score":0.5774000287055969},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.45719999074935913},{"id":"https://openalex.org/C3826847","wikidata":"https://www.wikidata.org/wiki/Q188768","display_name":"FLOPS","level":2,"score":0.4359999895095825},{"id":"https://openalex.org/C2986314615","wikidata":"https://www.wikidata.org/wiki/Q36829","display_name":"Pareto optimal","level":3,"score":0.41670000553131104},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.4106999933719635},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.37369999289512634},{"id":"https://openalex.org/C187191949","wikidata":"https://www.wikidata.org/wiki/Q1138496","display_name":"Profiling (computer programming)","level":2,"score":0.36340001225471497},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.36090001463890076},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.35429999232292175},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.3465999960899353},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.33889999985694885},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.3276999890804291},{"id":"https://openalex.org/C9819579","wikidata":"https://www.wikidata.org/wiki/Q1544018","display_name":"Tuner","level":3,"score":0.3215999901294708},{"id":"https://openalex.org/C205711294","wikidata":"https://www.wikidata.org/wiki/Q176953","display_name":"Rendering (computer graphics)","level":2,"score":0.3165000081062317},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.28540000319480896},{"id":"https://openalex.org/C199833920","wikidata":"https://www.wikidata.org/wiki/Q612536","display_name":"Vector quantization","level":2,"score":0.2824999988079071},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.27300000190734863},{"id":"https://openalex.org/C49154492","wikidata":"https://www.wikidata.org/wiki/Q5300","display_name":"Central processing unit","level":2,"score":0.26660001277923584},{"id":"https://openalex.org/C2944601119","wikidata":"https://www.wikidata.org/wiki/Q43744058","display_name":"Residual neural network","level":3,"score":0.2630000114440918},{"id":"https://openalex.org/C2989134064","wikidata":"https://www.wikidata.org/wiki/Q288510","display_name":"Execution time","level":2,"score":0.2623000144958496}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.1145/3786582.3786800","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3786582.3786800","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the IEEE/ACM 48th International Conference on Software Engineering","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2507.12196","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2507.12196","pdf_url":"https://arxiv.org/pdf/2507.12196","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,"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":null,"raw_type":"text"},{"id":"pmh:doi:10.48550/arxiv.2507.12196","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Article"},{"id":"doi:10.48550/arxiv.2507.12196","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2507.12196","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,"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":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.1145/3786582.3786800","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3786582.3786800","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the IEEE/ACM 48th International Conference on Software Engineering","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320322183","display_name":"Huawei Technologies","ror":"https://ror.org/00cmhce21"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Quantization":[0],"reduces":[1],"the":[2,18,60],"precision":[3],"of":[4,20,86,132,173],"deep":[5],"neural":[6],"networks":[7],"to":[8,125,163,171,177],"lower":[9,165],"model":[10],"size":[11,174],"and":[12,30,65,84,92,98,118,129,147],"computational":[13],"demands,":[14],"but":[15],"often":[16],"at":[17,55],"expense":[19],"accuracy.":[21],"Fully":[22],"quantized":[23,38,51,104,159,179],"models":[24,88,140],"can":[25],"suffer":[26],"significant":[27],"accuracy":[28,64,117,166],"degradation,":[29],"resource-constrained":[31],"hardware":[32,109],"accelerators":[33],"may":[34],"not":[35],"support":[36],"all":[37],"operations.":[39],"A":[40],"common":[41],"workaround":[42],"is":[43,67],"selective":[44,81],"quantization,":[45,82],"where":[46],"only":[47],"some":[48],"layers":[49],"are":[50],"while":[52,168],"others":[53],"remain":[54],"full":[56],"precision.":[57],"However,":[58],"determining":[59],"optimal":[61,127],"balance":[62],"between":[63],"efficiency":[66],"a":[68,77],"challenging":[69],"task.":[70],"To":[71],"this":[72],"direction,":[73],"we":[74],"propose":[75],"SeQTO,":[76],"framework":[78],"that":[79,153],"enables":[80],"deployment,":[83],"execution":[85],"ONNX":[87,139],"on":[89,113,137],"diverse":[90],"CPU":[91,146],"GPU":[93,148],"devices,":[94],"combined":[95],"with":[96],"profiling":[97],"multi-objective":[99],"optimization.":[100],"SeQTO":[101,136,154],"generates":[102],"selectively":[103,158],"models,":[105,160],"deploys":[106],"them":[107],"across":[108,145],"accelerators,":[110],"evaluates":[111],"performance":[112],"metrics":[114],"such":[115],"as":[116],"size,":[119],"applies":[120],"Pareto":[121],"Front-based":[122],"objective":[123],"minimization":[124],"identify":[126],"candidates,":[128],"provides":[130],"visualization":[131],"results.":[133],"We":[134],"evaluated":[135],"four":[138],"under":[141],"two":[142],"quantization":[143],"settings":[144],"devices.":[149],"Our":[150],"results":[151],"show":[152],"effectively":[155],"identifies":[156],"high-quality":[157],"achieving":[161],"up":[162,170],"54.14%":[164],"loss":[167],"maintaining":[169],"98.18%":[172],"reduction":[175],"compared":[176],"fully":[178],"models.":[180]},"counts_by_year":[],"updated_date":"2026-07-31T08:31:51.225901","created_date":"2025-10-14T00:00:00"}