{"id":"https://openalex.org/W7166881448","doi":"https://doi.org/10.18653/v1/2026.acl-long.1971","title":"ACBQ: Adaptive Cross-Block Quantization of Large Language Models","display_name":"ACBQ: Adaptive Cross-Block Quantization of Large Language Models","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166881448","doi":"https://doi.org/10.18653/v1/2026.acl-long.1971"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.acl-long.1971","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1971","pdf_url":"https://aclanthology.org/2026.acl-long.1971.pdf","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 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.acl-long.1971.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139841315","display_name":"Hailing Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I87182695","display_name":"Universidad del Noreste","ror":"https://ror.org/02ahky613","country_code":"MX","type":"education","lineage":["https://openalex.org/I87182695"]}],"countries":["MX"],"is_corresponding":false,"raw_author_name":"Hailing Wang","raw_affiliation_strings":["Northeastern University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Northeastern University","institution_ids":["https://openalex.org/I87182695"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139830895","display_name":"Jianglin Lu","orcid":null},"institutions":[{"id":"https://openalex.org/I87182695","display_name":"Universidad del Noreste","ror":"https://ror.org/02ahky613","country_code":"MX","type":"education","lineage":["https://openalex.org/I87182695"]}],"countries":["MX"],"is_corresponding":false,"raw_author_name":"Jianglin Lu","raw_affiliation_strings":["Northeastern University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Northeastern University","institution_ids":["https://openalex.org/I87182695"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127813716","display_name":"Yitian Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I87182695","display_name":"Universidad del Noreste","ror":"https://ror.org/02ahky613","country_code":"MX","type":"education","lineage":["https://openalex.org/I87182695"]}],"countries":["MX"],"is_corresponding":false,"raw_author_name":"Yitian Zhang","raw_affiliation_strings":["Northeastern University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Northeastern University","institution_ids":["https://openalex.org/I87182695"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139761638","display_name":"Huimin Zeng","orcid":null},"institutions":[{"id":"https://openalex.org/I87182695","display_name":"Universidad del Noreste","ror":"https://ror.org/02ahky613","country_code":"MX","type":"education","lineage":["https://openalex.org/I87182695"]}],"countries":["MX"],"is_corresponding":false,"raw_author_name":"Huimin Zeng","raw_affiliation_strings":["Northeastern University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Northeastern University","institution_ids":["https://openalex.org/I87182695"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5139758560","display_name":"Yun Fu","orcid":null},"institutions":[{"id":"https://openalex.org/I87182695","display_name":"Universidad del Noreste","ror":"https://ror.org/02ahky613","country_code":"MX","type":"education","lineage":["https://openalex.org/I87182695"]}],"countries":["MX"],"is_corresponding":false,"raw_author_name":"Yun Fu","raw_affiliation_strings":["Northeastern University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Northeastern University","institution_ids":["https://openalex.org/I87182695"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I87182695"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.86995957,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"42578","last_page":"42592"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.4740000069141388,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.4740000069141388,"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":0.15129999816417694,"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/T10028","display_name":"Topic Modeling","score":0.0617000013589859,"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":0.4675999879837036},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.37049999833106995},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.32330000400543213},{"id":"https://openalex.org/keywords/vector-quantization","display_name":"Vector quantization","score":0.28299999237060547}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5831999778747559},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4677000045776367},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.4675999879837036},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4113999903202057},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.37049999833106995},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.32330000400543213},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3010999858379364},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2964000105857849},{"id":"https://openalex.org/C199833920","wikidata":"https://www.wikidata.org/wiki/Q612536","display_name":"Vector quantization","level":2,"score":0.28299999237060547},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.28060001134872437}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.acl-long.1971","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1971","pdf_url":"https://aclanthology.org/2026.acl-long.1971.pdf","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 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.acl-long.1971","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1971","pdf_url":"https://aclanthology.org/2026.acl-long.1971.pdf","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 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166881448.pdf","grobid_xml":"https://content.openalex.org/works/W7166881448.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Post-training":[0],"quantization":[1,38,59,88,97,106,118,127],"(PTQ)":[2],"has":[3],"emerged":[4],"as":[5,104],"a":[6,78,95],"promising":[7],"approach":[8],"for":[9,132],"reducing":[10],"the":[11,48,113,147],"memory":[12],"footprint":[13],"and":[14,39,52,63,89,102,115,146,159],"computational":[15,168],"cost":[16],"of":[17,50,117],"large":[18],"language":[19],"models":[20],"(LLMs),":[21],"enabling":[22],"efficient":[23],"deployment":[24],"without":[25],"full":[26],"model":[27],"retraining.However,":[28],"existing":[29],"PTQ":[30],"methods":[31],"struggle":[32],"to":[33,61,69],"simultaneously":[34,84],"support":[35],"weight-activation":[36,86],"joint":[37,87],"extreme":[40,90],"low-bit":[41],"weight":[42,91],"quantization.This":[43],"limitation":[44],"primarily":[45],"arises":[46],"from":[47],"depth":[49],"LLMs":[51],"their":[53],"strong":[54],"cross-layer":[55],"dependencies,":[56],"which":[57],"cause":[58],"errors":[60,119],"propagate":[62],"accumulate":[64],"across":[65,120,138,141],"layers,":[66,121],"ultimately":[67],"leading":[68],"significant":[70],"performance":[71,155],"degradation.In":[72],"this":[73],"paper,":[74],"we":[75,122],"present":[76],"ACBQ,":[77],"simple":[79],"yet":[80],"effective":[81],"framework":[82],"that":[83,99,129,151],"addresses":[85],"quantization.We":[92],"first":[93],"propose":[94],"granular":[96],"strategy":[98,128],"treats":[100],"self-attention":[101],"FFN":[103],"separate":[105],"units":[107],"with":[108],"module-specific":[109],"optimization":[110],"objectives.To":[111],"mitigate":[112],"propagation":[114],"accumulation":[116],"introduce":[123],"an":[124],"adaptive":[125],"cross-block":[126],"explicitly":[130],"accounts":[131],"crosslayer":[133],"dependencies":[134],"by":[135],"encouraging":[136],"consistency":[137],"blocks.Extensive":[139],"experiments":[140],"diverse":[142],"LLMs,":[143],"including":[144],"OPT":[145],"LLaMA":[148],"family,":[149],"demonstrate":[150],"ACBQ":[152],"achieves":[153],"superior":[154],"under":[156],"both":[157],"W4A4":[158],"highly":[160],"aggressive":[161],"W2":[162],"settings,":[163],"while":[164],"incurring":[165],"negligible":[166],"additional":[167],"overhead.":[169]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}