{"id":"https://openalex.org/W4404133703","doi":"https://doi.org/10.1145/3649329.3656221","title":"Oltron: Algorithm-Hardware Co-design for Outlier-Aware Quantization of LLMs with Inter-/Intra-Layer Adaptation","display_name":"Oltron: Algorithm-Hardware Co-design for Outlier-Aware Quantization of LLMs with Inter-/Intra-Layer Adaptation","publication_year":2024,"publication_date":"2024-06-23","ids":{"openalex":"https://openalex.org/W4404133703","doi":"https://doi.org/10.1145/3649329.3656221"},"language":"en","primary_location":{"id":"doi:10.1145/3649329.3656221","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3649329.3656221","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 61st ACM/IEEE Design Automation Conference","raw_type":"proceedings-article"},"type":"article","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/A5104100813","display_name":"Chenhao Xue","orcid":"https://orcid.org/0000-0001-9721-247X"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]},{"id":"https://openalex.org/I4210165198","display_name":"Beijing Advanced Sciences and Innovation Center","ror":"https://ror.org/05qm21180","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165198"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Chenhao Xue","raw_affiliation_strings":["Beijing Advanced Innovation Center for Integrated Circuits, Beijing, Beijing, China","School of Integrated Circuits, Peking University, Beijing, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Beijing Advanced Innovation Center for Integrated Circuits, Beijing, Beijing, China","institution_ids":["https://openalex.org/I4210165198"]},{"raw_affiliation_string":"School of Integrated Circuits, Peking University, Beijing, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100374175","display_name":"Chen Zhang","orcid":"https://orcid.org/0000-0003-2762-2726"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chen Zhang","raw_affiliation_strings":["Shanghai Jiao Tong University, Shanghai, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University, Shanghai, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102441621","display_name":"Xun Jiang","orcid":"https://orcid.org/0009-0006-7958-8485"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xun Jiang","raw_affiliation_strings":["School of Integrated Circuits, Peking University, Beijing, Beijing, China"],"affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Peking University, Beijing, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082612170","display_name":"Zhen-Ran Gao","orcid":"https://orcid.org/0009-0008-1013-5781"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhutianya Gao","raw_affiliation_strings":["Shanghai Jiao Tong University, ShangHai, ShangHai, China"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University, ShangHai, ShangHai, China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000933188","display_name":"Yibo Lin","orcid":"https://orcid.org/0000-0002-0977-2774"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]},{"id":"https://openalex.org/I4210165198","display_name":"Beijing Advanced Sciences and Innovation Center","ror":"https://ror.org/05qm21180","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165198"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yibo Lin","raw_affiliation_strings":["Beijing Advanced Innovation Center for Integrated Circuits, Beijing, China","Institute of Electronic Design Automation, Peking University, Wuxi, China","School of Integrated Circuits, Peking University, Beijing, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Beijing Advanced Innovation Center for Integrated Circuits, Beijing, China","institution_ids":["https://openalex.org/I4210165198"]},{"raw_affiliation_string":"Institute of Electronic Design Automation, Peking University, Wuxi, China","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"School of Integrated Circuits, Peking University, Beijing, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5089719475","display_name":"Guangyu Sun","orcid":"https://orcid.org/0000-0002-7315-6589"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]},{"id":"https://openalex.org/I4210165198","display_name":"Beijing Advanced Sciences and Innovation Center","ror":"https://ror.org/05qm21180","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165198"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Guangyu Sun","raw_affiliation_strings":["Beijing Advanced Innovation Center for Integrated Circuits, Beijing, China","Institute of Electronic Design Automation, Peking University, Wuxi, China","School of Integrated Circuits, Peking University, Beijing, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Beijing Advanced Innovation Center for Integrated Circuits, Beijing, China","institution_ids":["https://openalex.org/I4210165198"]},{"raw_affiliation_string":"Institute of Electronic Design Automation, Peking University, Wuxi, China","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"School of Integrated Circuits, Peking University, Beijing, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5104100813"],"corresponding_institution_ids":["https://openalex.org/I20231570","https://openalex.org/I4210165198"],"apc_list":null,"apc_paid":null,"fwci":1.4614,"has_fulltext":false,"cited_by_count":7,"citation_normalized_percentile":{"value":0.82712367,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":98,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11338","display_name":"Advancements in Photolithography Techniques","score":0.9930999875068665,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic 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/T11338","display_name":"Advancements in Photolithography Techniques","score":0.9930999875068665,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11181","display_name":"Advanced Data Storage Technologies","score":0.9711999893188477,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T10558","display_name":"Advancements in Semiconductor Devices and Circuit Design","score":0.9391999840736389,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic 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":0.6660981178283691},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6474813222885132},{"id":"https://openalex.org/keywords/outlier","display_name":"Outlier","score":0.6181967258453369},{"id":"https://openalex.org/keywords/adaptation","display_name":"Adaptation (eye)","score":0.5803824663162231},{"id":"https://openalex.org/keywords/layer","display_name":"Layer (electronics)","score":0.46773844957351685},{"id":"https://openalex.org/keywords/algorithm-design","display_name":"Algorithm design","score":0.44401827454566956},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.42307600378990173},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.21436229348182678},{"id":"https://openalex.org/keywords/psychology","display_name":"Psychology","score":0.10669052600860596},{"id":"https://openalex.org/keywords/nanotechnology","display_name":"Nanotechnology","score":0.06240853667259216},{"id":"https://openalex.org/keywords/materials-science","display_name":"Materials science","score":0.057620078325271606}],"concepts":[{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.6660981178283691},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6474813222885132},{"id":"https://openalex.org/C79337645","wikidata":"https://www.wikidata.org/wiki/Q779824","display_name":"Outlier","level":2,"score":0.6181967258453369},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.5803824663162231},{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.46773844957351685},{"id":"https://openalex.org/C106516650","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm design","level":2,"score":0.44401827454566956},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.42307600378990173},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.21436229348182678},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.10669052600860596},{"id":"https://openalex.org/C171250308","wikidata":"https://www.wikidata.org/wiki/Q11468","display_name":"Nanotechnology","level":1,"score":0.06240853667259216},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.057620078325271606},{"id":"https://openalex.org/C169760540","wikidata":"https://www.wikidata.org/wiki/Q207011","display_name":"Neuroscience","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3649329.3656221","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3649329.3656221","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 61st ACM/IEEE Design Automation Conference","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":13,"referenced_works":["https://openalex.org/W2072730350","https://openalex.org/W2883920103","https://openalex.org/W2963367920","https://openalex.org/W2981852735","https://openalex.org/W3030163527","https://openalex.org/W3100985894","https://openalex.org/W3133253223","https://openalex.org/W4281651027","https://openalex.org/W4288089799","https://openalex.org/W4292779060","https://openalex.org/W4308083739","https://openalex.org/W4366341968","https://openalex.org/W4385245566"],"related_works":["https://openalex.org/W3006513224","https://openalex.org/W2046456988","https://openalex.org/W2357409937","https://openalex.org/W2510582230","https://openalex.org/W2978674666","https://openalex.org/W2074430941","https://openalex.org/W2113096305","https://openalex.org/W2997567050","https://openalex.org/W1977636359","https://openalex.org/W2772305933"],"abstract_inverted_index":{"In":[0],"Large":[1],"Language":[2],"Models":[3],"(LLMs),":[4],"outliers":[5,60,102],"are":[6],"identified":[7],"by":[8],"a":[9,72,92,110,113,125,148],"small":[10],"number":[11],"of":[12,61,80,101],"values":[13],"with":[14,46,82],"exceptionally":[15],"high":[16],"magnitudes,":[17],"critically":[18],"affecting":[19],"model":[20,44,160],"accuracy.":[21,161],"Researchers":[22],"have":[23],"proposed":[24],"several":[25],"mixed-precision":[26,138],"quantization":[27,79,94],"techniques":[28],"to":[29,118],"manage":[30],"these":[31],"activation":[32,59],"outliers.":[33],"These":[34],"approaches,":[35],"employing":[36],"value-wise":[37],"outlier":[38],"granularity,":[39],"face":[40],"challenges":[41],"in":[42],"balancing":[43],"accuracy":[45],"hardware":[47],"efficiency.":[48],"To":[49],"address":[50],"this":[51],"issue,":[52],"we":[53,69],"capitalize":[54],"on":[55],"the":[56,98,142],"observation":[57],"that":[58,96,116,129],"LLMs":[62,81],"typically":[63],"cluster":[64],"within":[65,109],"specific":[66],"channels.":[67],"Consequently,":[68],"introduce":[70],"Oltron,":[71],"comprehensive":[73],"software/hardware":[74],"co-design":[75],"strategy":[76],"for":[77,137],"outlier-aware":[78,144],"inter-/intra-layer":[83],"adaptation.":[84],"Our":[85],"method":[86],"includes":[87],"three":[88],"key":[89],"innovations:":[90],"firstly,":[91],"novel":[93],"algorithm":[95],"identifies":[97],"optimal":[99],"ratio":[100],"across":[103],"different":[104],"layers":[105],"and":[106,120,123,134,152],"channel":[107],"groups":[108],"layer;":[111],"secondly,":[112],"reconfigurable":[114],"architecture":[115],"adapts":[117],"inter-":[119],"intra-layer":[121],"distributions;":[122],"thirdly,":[124],"tile-based":[126],"dataflow":[127],"optimizer":[128],"intricately":[130],"arranges":[131],"complex":[132],"computations":[133],"memory":[135],"access":[136],"tensors.":[139],"Oltron":[140],"outperforms":[141],"state-of-the-art":[143],"accelerator,":[145],"OliVe,":[146],"achieving":[147],"1.9x":[149],"performance":[150],"boost":[151],"1.6x":[153],"greater":[154],"energy":[155],"efficiency,":[156],"while":[157],"also":[158],"enhancing":[159]},"counts_by_year":[{"year":2025,"cited_by_count":7}],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}
