{"id":"https://openalex.org/W4408860980","doi":"https://doi.org/10.1109/icce63647.2025.10930198","title":"Selective Error Correction for Activation in Large Language Model Training","display_name":"Selective Error Correction for Activation in Large Language Model Training","publication_year":2025,"publication_date":"2025-01-11","ids":{"openalex":"https://openalex.org/W4408860980","doi":"https://doi.org/10.1109/icce63647.2025.10930198"},"language":"en","primary_location":{"id":"doi:10.1109/icce63647.2025.10930198","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icce63647.2025.10930198","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Conference on Consumer Electronics (ICCE)","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/A5101408508","display_name":"Seungyong Lee","orcid":"https://orcid.org/0000-0002-7348-5454"},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":true,"raw_author_name":"Seungyong Lee","raw_affiliation_strings":["Seoul National University,Department of Electrical and Computer Engineering,Seoul,Korea"],"affiliations":[{"raw_affiliation_string":"Seoul National University,Department of Electrical and Computer Engineering,Seoul,Korea","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5102861073","display_name":"Hyuk-Jae Lee","orcid":"https://orcid.org/0000-0001-8895-9117"},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Hyuk-Jae Lee","raw_affiliation_strings":["Seoul National University,Department of Electrical and Computer Engineering,Seoul,Korea"],"affiliations":[{"raw_affiliation_string":"Seoul National University,Department of Electrical and Computer Engineering,Seoul,Korea","institution_ids":["https://openalex.org/I139264467"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5101408508"],"corresponding_institution_ids":["https://openalex.org/I139264467"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.02451703,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"2"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9437000155448914,"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":0.9437000155448914,"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/computer-science","display_name":"Computer science","score":0.7544211149215698},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5483744740486145},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.5145799517631531},{"id":"https://openalex.org/keywords/error-detection-and-correction","display_name":"Error detection and correction","score":0.49927473068237305},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.49201056361198425},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.44764021039009094},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.41805294156074524},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.1879909634590149}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7544211149215698},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5483744740486145},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.5145799517631531},{"id":"https://openalex.org/C103088060","wikidata":"https://www.wikidata.org/wiki/Q1062839","display_name":"Error detection and correction","level":2,"score":0.49927473068237305},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.49201056361198425},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.44764021039009094},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.41805294156074524},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.1879909634590149},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icce63647.2025.10930198","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icce63647.2025.10930198","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Conference on Consumer Electronics (ICCE)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.44999998807907104}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":3,"referenced_works":["https://openalex.org/W2981860227","https://openalex.org/W4280488201","https://openalex.org/W4392943848"],"related_works":["https://openalex.org/W230091440","https://openalex.org/W2233261550","https://openalex.org/W2810751659","https://openalex.org/W258997015","https://openalex.org/W2997094352","https://openalex.org/W3216976533","https://openalex.org/W100620283","https://openalex.org/W2495260952","https://openalex.org/W4366179611","https://openalex.org/W2996078371"],"abstract_inverted_index":{"In":[0,25],"Large":[1],"Language":[2],"Model":[3],"(LLM)":[4],"training,":[5],"acti-vations":[6],"constitute":[7],"a":[8,30,58],"significant":[9,61],"portion":[10],"of":[11,41,52,60],"memory":[12],"usage,":[13],"and":[14,74,99],"memory-side":[15],"errors":[16],"occurring":[17],"in":[18,88,109],"activations":[19,53],"can":[20],"lead":[21],"to":[22,37,81,85],"training":[23,107],"failures.":[24],"this":[26],"paper,":[27],"we":[28],"propose":[29],"lightweight":[31],"but":[32],"effective":[33],"error":[34,50,67],"correction":[35],"method":[36,92],"mitigate":[38],"the":[39,48,90],"impact":[40],"such":[42],"errors.":[43],"The":[44],"proposed":[45,91],"approach":[46],"leverages":[47],"inherent":[49],"tolerance":[51],"by":[54],"selectively":[55],"protecting":[56],"only":[57],"subset":[59],"bits.":[62,83],"Specifically,":[63],"based":[64],"on":[65],"bit-flip":[66],"analysis,":[68],"six":[69],"sensitive":[70],"bits":[71],"are":[72,78],"identified,":[73],"error-correcting":[75],"codes":[76],"(ECC)":[77],"applied":[79],"exclusively":[80],"these":[82],"Compared":[84],"ECC":[86],"implemented":[87],"HBM2,":[89],"exhibits":[93],"5.4":[94],"times":[95,101],"lower":[96],"parity":[97],"overhead":[98],"4":[100],"higher":[102],"area":[103],"efficiency,":[104],"while":[105],"maintaining":[106],"performance":[108],"error-prone":[110],"environments.":[111]},"counts_by_year":[],"updated_date":"2025-12-28T23:10:05.387466","created_date":"2025-10-10T00:00:00"}