{"id":"https://openalex.org/W7165115163","doi":"https://doi.org/10.1145/3787109.3815266","title":"ReCQ: Residual Compensation for Quantized Convolutional Neural Networks","display_name":"ReCQ: Residual Compensation for Quantized Convolutional Neural Networks","publication_year":2026,"publication_date":"2026-06-18","ids":{"openalex":"https://openalex.org/W7165115163","doi":"https://doi.org/10.1145/3787109.3815266"},"language":null,"primary_location":{"id":"doi:10.1145/3787109.3815266","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3787109.3815266","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 Great Lakes Symposium on VLSI 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3787109.3815266","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5053708213","display_name":"Mohammadreza Mohammadi","orcid":"https://orcid.org/0000-0002-1471-9871"},"institutions":[{"id":"https://openalex.org/I155781252","display_name":"University of South Carolina","ror":"https://ror.org/02b6qw903","country_code":"US","type":"education","lineage":["https://openalex.org/I155781252"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mohammadreza Mohammadi","raw_affiliation_strings":["Computer Science and Engineering, University of South Carolina, Columbia, SC, USA"],"raw_orcid":"https://orcid.org/0000-0002-1471-9871","affiliations":[{"raw_affiliation_string":"Computer Science and Engineering, University of South Carolina, Columbia, SC, USA","institution_ids":["https://openalex.org/I155781252"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119386729","display_name":"Matthew Grenier","orcid":null},"institutions":[{"id":"https://openalex.org/I155781252","display_name":"University of South Carolina","ror":"https://ror.org/02b6qw903","country_code":"US","type":"education","lineage":["https://openalex.org/I155781252"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Matthew Grenier","raw_affiliation_strings":["Computer Science and Engineering, University of South Carolina, Columbia, SC, USA"],"raw_orcid":"https://orcid.org/0009-0009-5785-086X","affiliations":[{"raw_affiliation_string":"Computer Science and Engineering, University of South Carolina, Columbia, SC, USA","institution_ids":["https://openalex.org/I155781252"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5115597960","display_name":"Ramtin Zand","orcid":"https://orcid.org/0000-0002-1786-1152"},"institutions":[{"id":"https://openalex.org/I155781252","display_name":"University of South Carolina","ror":"https://ror.org/02b6qw903","country_code":"US","type":"education","lineage":["https://openalex.org/I155781252"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ramtin Zand","raw_affiliation_strings":["Computer Science and Engineering, University of South Carolina, Columbia, SC, USA"],"raw_orcid":"https://orcid.org/0000-0002-1786-1152","affiliations":[{"raw_affiliation_string":"Computer Science and Engineering, University of South Carolina, Columbia, SC, USA","institution_ids":["https://openalex.org/I155781252"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I155781252"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.76648579,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"543","last_page":"548"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9343000054359436,"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":0.9343000054359436,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.014999999664723873,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.012900000438094139,"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/convolutional-neural-network","display_name":"Convolutional neural network","score":0.7993000149726868},{"id":"https://openalex.org/keywords/residual","display_name":"Residual","score":0.7961000204086304},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.6820999979972839},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.4936000108718872},{"id":"https://openalex.org/keywords/compensation","display_name":"Compensation (psychology)","score":0.4392000138759613},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.38749998807907104}],"concepts":[{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.7993000149726868},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.7961000204086304},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7373999953269958},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.6820999979972839},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5321999788284302},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4936000108718872},{"id":"https://openalex.org/C2780023022","wikidata":"https://www.wikidata.org/wiki/Q1338171","display_name":"Compensation (psychology)","level":2,"score":0.4392000138759613},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3912999927997589},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.38749998807907104},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.37279999256134033},{"id":"https://openalex.org/C157899210","wikidata":"https://www.wikidata.org/wiki/Q1395022","display_name":"Convolutional code","level":3,"score":0.35850000381469727},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3028999865055084},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.29829999804496765},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2806999981403351},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.25220000743865967}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3787109.3815266","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3787109.3815266","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 Great Lakes Symposium on VLSI 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3787109.3815266","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3787109.3815266","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 Great Lakes Symposium on VLSI 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1315800894","display_name":"CAREER: Heterogeneous Neuromorphic and Edge Computing Systems for Realtime Machine Learning Technologies","funder_award_id":"2340249","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G4428402310","display_name":"SHF: Small: Cross-Layer Design Automation for In-Memory Analog Computing","funder_award_id":"2409697","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":15,"referenced_works":["https://openalex.org/W2117539524","https://openalex.org/W2194775991","https://openalex.org/W2796438033","https://openalex.org/W2936252403","https://openalex.org/W2962298324","https://openalex.org/W3127215335","https://openalex.org/W3137147200","https://openalex.org/W4225374436","https://openalex.org/W4312443924","https://openalex.org/W4387965444","https://openalex.org/W4387968184","https://openalex.org/W4390704242","https://openalex.org/W4401726555","https://openalex.org/W4404212170","https://openalex.org/W4413144723"],"related_works":[],"abstract_inverted_index":{"Post-training":[0],"quantization":[1,25],"(PTQ)":[2],"is":[3],"widely":[4],"used":[5],"to":[6,49,130],"reduce":[7],"the":[8,80,124,134],"computational":[9],"and":[10,98,113,133,162],"memory":[11],"requirements":[12],"of":[13,91,126],"deep":[14,33],"neural":[15,55],"networks":[16,56],"for":[17],"deployment":[18],"on":[19],"resource-constrained":[20],"platforms.":[21],"However,":[22],"aggressive":[23],"low-bit":[24],"often":[26],"introduces":[27],"significant":[28],"accuracy":[29,125,151],"degradation,":[30],"particularly":[31],"in":[32,53,69],"convolutional":[34,54,73,107,115],"architectures.":[35,168],"To":[36],"address":[37],"this":[38],"challenge,":[39],"we":[40],"propose":[41],"ReCQ,":[42],"a":[43,88],"lightweight":[44],"residual":[45,76,109],"compensation":[46,65],"framework":[47],"designed":[48],"recover":[50],"quantization-induced":[51],"errors":[52],"(CNNs).":[57],"ReCQ":[58,86,121,148],"augments":[59],"quantized":[60,127],"models":[61,128],"with":[62,71,152],"small":[63],"depthwise-separable":[64,111],"modules":[66],"that":[67,120],"operate":[68],"parallel":[70],"existing":[72],"blocks,":[74],"learning":[75],"corrections":[77],"while":[78,138],"preserving":[79],"original":[81],"network":[82],"structure.":[83],"We":[84],"evaluate":[85],"across":[87,165],"diverse":[89],"set":[90],"CNN":[92,167],"architectures,":[93,112],"including":[94],"VGG,":[95],"ResNet,":[96],"MobileNetV2,":[97],"ConvNeXt,":[99],"representing":[100],"different":[101],"design":[102],"paradigms":[103],"such":[104],"as":[105],"plain":[106],"networks,":[108,110],"modern":[114,166],"backbones.":[116],"Experimental":[117],"results":[118],"show":[119],"consistently":[122],"improves":[123],"compared":[129],"standard":[131],"PTQ":[132],"recent":[135],"QwT":[136],"method":[137],"introducing":[139],"only":[140],"minimal":[141],"model":[142],"size":[143],"overhead.":[144],"In":[145],"many":[146],"configurations,":[147],"achieves":[149],"higher":[150],"comparable":[153],"or":[154],"even":[155],"smaller":[156],"parameter":[157],"overhead,":[158],"demonstrating":[159],"its":[160],"effectiveness":[161],"general":[163],"applicability":[164]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-06-19T00:00:00"}