{"id":"https://openalex.org/W4391410261","doi":"https://doi.org/10.1109/mwscas57524.2023.10406106","title":"Hessian-Based Parameter Quantization Method for BERT","display_name":"Hessian-Based Parameter Quantization Method for BERT","publication_year":2023,"publication_date":"2023-08-06","ids":{"openalex":"https://openalex.org/W4391410261","doi":"https://doi.org/10.1109/mwscas57524.2023.10406106"},"language":"en","primary_location":{"id":"doi:10.1109/mwscas57524.2023.10406106","is_oa":false,"landing_page_url":"https://doi.org/10.1109/mwscas57524.2023.10406106","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE 66th International Midwest Symposium on Circuits and Systems (MWSCAS)","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/A5109786813","display_name":"Woohong Byun","orcid":null},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Woohong Byun","raw_affiliation_strings":["School of Electrical and Computer Engineering, Georgia Institute of Technology,Atlanta,GA,30332"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Electrical and Computer Engineering, Georgia Institute of Technology,Atlanta,GA,30332","institution_ids":["https://openalex.org/I130701444"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5009591041","display_name":"Saibal Mukhopadhyay","orcid":"https://orcid.org/0000-0002-8894-3390"},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Saibal Mukhopadhyay","raw_affiliation_strings":["School of Electrical and Computer Engineering, Georgia Institute of Technology,Atlanta,GA,30332"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Electrical and Computer Engineering, Georgia Institute of Technology,Atlanta,GA,30332","institution_ids":["https://openalex.org/I130701444"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I130701444"],"apc_list":null,"apc_paid":null,"fwci":0.8712,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.77091119,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"516","last_page":"520"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9998000264167786,"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":0.9998000264167786,"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":0.9994999766349792,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9954000115394592,"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":0.7576872110366821},{"id":"https://openalex.org/keywords/hessian-matrix","display_name":"Hessian matrix","score":0.6703657507896423},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6567438244819641},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.619172215461731},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.4547592103481293},{"id":"https://openalex.org/keywords/metric","display_name":"Metric (unit)","score":0.4178413450717926},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.38224977254867554},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.21294626593589783},{"id":"https://openalex.org/keywords/applied-mathematics","display_name":"Applied mathematics","score":0.10671982169151306},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.057889461517333984},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.05741521716117859}],"concepts":[{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.7576872110366821},{"id":"https://openalex.org/C203616005","wikidata":"https://www.wikidata.org/wiki/Q620495","display_name":"Hessian matrix","level":2,"score":0.6703657507896423},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6567438244819641},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.619172215461731},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4547592103481293},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.4178413450717926},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.38224977254867554},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.21294626593589783},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.10671982169151306},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.057889461517333984},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.05741521716117859},{"id":"https://openalex.org/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","level":1,"score":0.0},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/mwscas57524.2023.10406106","is_oa":false,"landing_page_url":"https://doi.org/10.1109/mwscas57524.2023.10406106","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE 66th International Midwest Symposium on Circuits and Systems (MWSCAS)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":21,"referenced_works":["https://openalex.org/W2590016403","https://openalex.org/W2613543507","https://openalex.org/W2735441098","https://openalex.org/W2896457183","https://openalex.org/W2923014074","https://openalex.org/W2980965328","https://openalex.org/W2996428491","https://openalex.org/W2998183051","https://openalex.org/W3033529678","https://openalex.org/W3098576111","https://openalex.org/W3100985894","https://openalex.org/W3105966348","https://openalex.org/W3177265267","https://openalex.org/W4236601256","https://openalex.org/W6755207826","https://openalex.org/W6768021236","https://openalex.org/W6768086466","https://openalex.org/W6768902189","https://openalex.org/W6779163297","https://openalex.org/W6784041521","https://openalex.org/W6786404238"],"related_works":["https://openalex.org/W2611031068","https://openalex.org/W1704347466","https://openalex.org/W4283017538","https://openalex.org/W1996936972","https://openalex.org/W1545275724","https://openalex.org/W2802707792","https://openalex.org/W2569979269","https://openalex.org/W2075777916","https://openalex.org/W3021699548","https://openalex.org/W4291718021"],"abstract_inverted_index":{"Transformer-based":[0],"language":[1],"models":[2],"have":[3],"shown":[4],"outstanding":[5],"performance":[6],"in":[7,76],"various":[8],"NLP":[9],"tasks,":[10],"but":[11],"using":[12,57],"them":[13],"on":[14,83],"edge":[15],"devices":[16],"is":[17],"very":[18],"challenging":[19],"due":[20],"to":[21,86],"their":[22],"notorious":[23],"memory":[24],"usage.":[25],"To":[26],"address":[27],"this":[28,30],"issue,":[29],"paper":[31],"proposes":[32],"a":[33,58,79],"novel":[34],"parameter":[35],"quantization":[36,96],"method":[37,68],"for":[38],"BERT":[39,77,88],"that":[40,66],"quantizes":[41],"important":[42],"parameters":[43,51,75],"with":[44,52,78],"higher":[45],"precision":[46,54],"bit":[47,55],"width":[48,56],"and":[49],"unimportant":[50],"lower":[53],"Hessian-based":[59],"sensitivity":[60],"metric.":[61],"The":[62],"experimental":[63],"results":[64],"show":[65],"our":[67],"achieves":[69],"19.6X":[70],"compression":[71],"of":[72],"the":[73,87],"model":[74],"0.8%":[80],"accuracy":[81],"drop":[82],"MNLI":[84],"compared":[85],"base":[89],"model,":[90],"generally":[91],"outperforming":[92],"other":[93],"existing":[94],"layer-wise":[95],"methods.":[97]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":2}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}