{"id":"https://openalex.org/W4402753294","doi":"https://doi.org/10.1109/cis-ram61939.2024.10672819","title":"GPTQT: Quantize Large Language Models Twice to Push the Efficiency","display_name":"GPTQT: Quantize Large Language Models Twice to Push the Efficiency","publication_year":2024,"publication_date":"2024-08-08","ids":{"openalex":"https://openalex.org/W4402753294","doi":"https://doi.org/10.1109/cis-ram61939.2024.10672819"},"language":"en","primary_location":{"id":"doi:10.1109/cis-ram61939.2024.10672819","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cis-ram61939.2024.10672819","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE International Conference on Cybernetics and Intelligent Systems (CIS) and IEEE International Conference on Robotics, Automation and Mechatronics (RAM)","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/A5031339858","display_name":"Yipin Guo","orcid":"https://orcid.org/0000-0003-4335-2826"},"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":"Yipin Guo","raw_affiliation_strings":["Zhejiang University,College of Control Science and Engineering,Hangzhou,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University,College of Control Science and Engineering,Hangzhou,China","institution_ids":["https://openalex.org/I76130692"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050210424","display_name":"Yilin Lang","orcid":"https://orcid.org/0000-0002-9381-9888"},"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":"Yilin Lang","raw_affiliation_strings":["Zhejiang University,College of Control Science and Engineering,Hangzhou,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University,College of Control Science and Engineering,Hangzhou,China","institution_ids":["https://openalex.org/I76130692"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5032554855","display_name":"Qinyuan Ren","orcid":null},"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":"Qinyuan Ren","raw_affiliation_strings":["Zhejiang University,College of Control Science and Engineering,Hangzhou,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University,College of Control Science and Engineering,Hangzhou,China","institution_ids":["https://openalex.org/I76130692"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I76130692"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"368","last_page":"373"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9984999895095825,"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/T10028","display_name":"Topic Modeling","score":0.9984999895095825,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9973000288009644,"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/T11719","display_name":"Data Quality and Management","score":0.982200026512146,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6909207105636597},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.3833644390106201}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6909207105636597},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3833644390106201}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/cis-ram61939.2024.10672819","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cis-ram61939.2024.10672819","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE International Conference on Cybernetics and Intelligent Systems (CIS) and IEEE International Conference on Robotics, Automation and Mechatronics (RAM)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":27,"referenced_works":["https://openalex.org/W2300242332","https://openalex.org/W2913668833","https://openalex.org/W2990138404","https://openalex.org/W3120384882","https://openalex.org/W4229005866","https://openalex.org/W4288374981","https://openalex.org/W4293166090","https://openalex.org/W4298422451","https://openalex.org/W4308760226","https://openalex.org/W4311642023","https://openalex.org/W4322718191","https://openalex.org/W4379260375","https://openalex.org/W4391670689","https://openalex.org/W4394828356","https://openalex.org/W6727099177","https://openalex.org/W6761783079","https://openalex.org/W6788514523","https://openalex.org/W6811340617","https://openalex.org/W6838633097","https://openalex.org/W6839092155","https://openalex.org/W6842795023","https://openalex.org/W6847478871","https://openalex.org/W6850625674","https://openalex.org/W6853048723","https://openalex.org/W6853251322","https://openalex.org/W6856931071","https://openalex.org/W6862355752"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052","https://openalex.org/W2382290278","https://openalex.org/W4395014643"],"abstract_inverted_index":{"Due":[0],"to":[1,25,54,69,80,90,118],"their":[2],"large":[3],"size,":[4],"generative":[5],"Large":[6],"Language":[7],"Models":[8],"(LLMs)":[9],"require":[10],"significant":[11],"computing":[12],"and":[13,29,112,132],"storage":[14],"resources.":[15],"This":[16],"paper":[17],"introduces":[18],"a":[19,59,70],"new":[20],"post-training":[21],"quantization":[22,47,68,122,152],"method,":[23],"GPTQT,":[24],"reduce":[26],"memory":[27],"usage":[28],"enhance":[30],"processing":[31],"speed":[32,134],"by":[33,75,128,135],"expressing":[34],"the":[35,46,119,148],"weight":[36,79],"of":[37,49,157],"LLM":[38],"in":[39],"3bit/2bit.":[40],"Practice":[41],"has":[42],"shown":[43],"that":[44,144],"minimizing":[45],"error":[48],"weights":[50,65],"is":[51,88,146],"ineffective,":[52],"leading":[53],"overfitting.":[55],"Therefore,":[56],"GPTQT":[57,124,145],"employs":[58],"progressive":[60],"two-step":[61],"approach:":[62],"initially":[63],"quantizing":[64],"using":[66],"Linear":[67],"relatively":[71],"high":[72],"bit,":[73],"followed":[74],"converting":[76],"obtained":[77],"int":[78],"lower":[81],"bit":[82],"binary":[83,103,150],"coding.":[84],"A":[85],"re-explore":[86],"strategy":[87],"proposed":[89],"optimize":[91],"initial":[92],"scaling":[93],"factor.":[94],"During":[95],"inference,":[96],"these":[97],"steps":[98],"are":[99],"merged":[100],"into":[101],"pure":[102],"coding,":[104],"enabling":[105],"efficient":[106],"computation.":[107],"Testing":[108],"across":[109],"various":[110],"models":[111],"datasets":[113],"confirms":[114],"GPTQT\u2019s":[115],"effectiveness.":[116],"Compared":[117],"strong":[120],"3-bit":[121],"baseline,":[123],"further":[125],"reduces":[126],"perplexity":[127],"4.01":[129],"on":[130,137,141],"opt-66B":[131],"increases":[133],"1.24\u00d7":[136],"opt-30b.":[138],"The":[139],"results":[140],"Llama2":[142],"show":[143],"currently":[147],"best":[149],"coding":[151],"method":[153],"for":[154],"such":[155],"kind":[156],"LLMs.":[158]},"counts_by_year":[{"year":2025,"cited_by_count":2}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}