{"id":"https://openalex.org/W7138247086","doi":"https://doi.org/10.1609/aaai.v40i26.39283","title":"FLRQ: Faster LLM Quantization with Flexible Low-Rank Matrix Sketching","display_name":"FLRQ: Faster LLM Quantization with Flexible Low-Rank Matrix Sketching","publication_year":2026,"publication_date":"2026-03-14","ids":{"openalex":"https://openalex.org/W7138247086","doi":"https://doi.org/10.1609/aaai.v40i26.39283"},"language":null,"primary_location":{"id":"doi:10.1609/aaai.v40i26.39283","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i26.39283","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://doi.org/10.1609/aaai.v40i26.39283","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5129647450","display_name":"Hongyaoxing Gu","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Hongyaoxing Gu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129701732","display_name":"Lijuan Hu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lijuan Hu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5067174475","display_name":"Shuzi Niu","orcid":"https://orcid.org/0000-0001-9648-1803"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shuzi Niu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5129657819","display_name":"Fangfang Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fangfang Liu","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5129647450"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.59060403,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"40","issue":"26","first_page":"21369","last_page":"21377"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.28519999980926514,"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":0.28519999980926514,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.14380000531673431,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.08060000091791153,"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/quantization","display_name":"Quantization (signal processing)","score":0.661300003528595},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.5597000122070312},{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.5034000277519226},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.46549999713897705},{"id":"https://openalex.org/keywords/clipping","display_name":"Clipping (morphology)","score":0.44850000739097595},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4212000072002411},{"id":"https://openalex.org/keywords/gramian-matrix","display_name":"Gramian matrix","score":0.41839998960494995},{"id":"https://openalex.org/keywords/gaussian","display_name":"Gaussian","score":0.41269999742507935}],"concepts":[{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.661300003528595},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.6068999767303467},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5723999738693237},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.5597000122070312},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.5034000277519226},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.46549999713897705},{"id":"https://openalex.org/C2776848632","wikidata":"https://www.wikidata.org/wiki/Q853463","display_name":"Clipping (morphology)","level":2,"score":0.44850000739097595},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4212000072002411},{"id":"https://openalex.org/C77246614","wikidata":"https://www.wikidata.org/wiki/Q1409400","display_name":"Gramian matrix","level":3,"score":0.41839998960494995},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.41269999742507935},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.4016999900341034},{"id":"https://openalex.org/C70451592","wikidata":"https://www.wikidata.org/wiki/Q376493","display_name":"Dither","level":3,"score":0.38989999890327454},{"id":"https://openalex.org/C84507854","wikidata":"https://www.wikidata.org/wiki/Q952222","display_name":"Diagonally dominant matrix","level":3,"score":0.3587999939918518},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3578000068664551},{"id":"https://openalex.org/C90199385","wikidata":"https://www.wikidata.org/wiki/Q6692777","display_name":"Low-rank approximation","level":3,"score":0.35690000653266907},{"id":"https://openalex.org/C2777036070","wikidata":"https://www.wikidata.org/wiki/Q18393452","display_name":"Random projection","level":2,"score":0.3010999858379364},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.29750001430511475},{"id":"https://openalex.org/C159694833","wikidata":"https://www.wikidata.org/wiki/Q2321565","display_name":"Iterative method","level":2,"score":0.2874000072479248},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.2759000062942505},{"id":"https://openalex.org/C167085575","wikidata":"https://www.wikidata.org/wiki/Q6803654","display_name":"Mean squared prediction error","level":2,"score":0.2721000015735626},{"id":"https://openalex.org/C148764684","wikidata":"https://www.wikidata.org/wiki/Q621751","display_name":"Approximation algorithm","level":2,"score":0.26260000467300415},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.2612999975681305},{"id":"https://openalex.org/C199833920","wikidata":"https://www.wikidata.org/wiki/Q612536","display_name":"Vector quantization","level":2,"score":0.25130000710487366}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1609/aaai.v40i26.39283","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i26.39283","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1609/aaai.v40i26.39283","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i26.39283","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Traditional":[0],"post-training":[1],"quantization":[2,144,170],"(PTQ)":[3],"is":[4],"considered":[5],"an":[6,153],"effective":[7],"approach":[8],"to":[9,29,43,87,96],"reduce":[10],"model":[11],"size":[12],"and":[13,37,93,111,149,160,172],"accelerate":[14],"inference":[15],"of":[16,67],"large-scale":[17],"language":[18],"models":[19],"(LLMs).":[20],"However,":[21],"existing":[22],"low-rank":[23,52,68,128,143],"PTQ":[24],"methods":[25],"require":[26],"costly":[27],"fine-tuning":[28],"determine":[30],"a":[31,83],"compromise":[32],"rank":[33,132],"for":[34,125,134],"diverse":[35],"data":[36],"layers":[38,72],"in":[39,73,162,168],"large":[40],"models,":[41],"failing":[42],"exploit":[44],"their":[45],"full":[46],"potential.":[47],"Additionally,":[48],"the":[49,55,64,90,120,126,142,147],"current":[50],"SVD-based":[51],"approximation":[53,69],"compounds":[54],"computational":[56],"overhead.":[57],"In":[58],"this":[59],"work,":[60],"we":[61,77],"thoroughly":[62],"analyze":[63],"varying":[65],"effectiveness":[66,159],"across":[70],"different":[71],"representative":[74],"models.":[75],"Accordingly,":[76],"introduce":[78],"Flexible":[79,107],"Low-Rank":[80],"Quantization":[81],"(FLRQ),":[82],"novel":[84],"solution":[85],"designed":[86],"quickly":[88],"identify":[89],"accuracy-optimal":[91],"ranks":[92],"aggregate":[94],"them":[95],"achieve":[97],"minimal":[98],"storage":[99],"combinations.":[100],"FLRQ":[101,156],"comprises":[102],"two":[103],"powerful":[104],"components,":[105],"Rank1-Sketch-based":[106],"Rank":[108],"Selection":[109],"(R1-FLR)":[110],"Best":[112],"Low-rank":[113],"Approximation":[114],"under":[115,146],"Clipping":[116],"(BLC).":[117],"R1-FLR":[118],"applies":[119],"R1-Sketch":[121],"with":[122],"Gaussian":[123],"projection":[124],"fast":[127],"approximation,":[129],"enabling":[130],"outlier-aware":[131],"extraction":[133],"each":[135],"layer.":[136],"Meanwhile,":[137],"BLC":[138],"aims":[139],"at":[140],"minimizing":[141],"error":[145],"scaling":[148],"clipping":[150],"strategy":[151],"through":[152],"iterative":[154],"method.":[155],"demonstrates":[157],"strong":[158],"robustness":[161],"comprehensive":[163],"experiments,":[164],"achieving":[165],"state-of-the-art":[166],"performance":[167],"both":[169],"quality":[171],"algorithm":[173],"efficiency.":[174]},"counts_by_year":[],"updated_date":"2026-03-18T06:31:55.123368","created_date":"2026-03-18T00:00:00"}