{"id":"https://openalex.org/W4408925916","doi":"https://doi.org/10.1145/3676641.3715998","title":"FlexSP: Accelerating Large Language Model Training via Flexible Sequence Parallelism","display_name":"FlexSP: Accelerating Large Language Model Training via Flexible Sequence Parallelism","publication_year":2025,"publication_date":"2025-03-27","ids":{"openalex":"https://openalex.org/W4408925916","doi":"https://doi.org/10.1145/3676641.3715998"},"language":"en","primary_location":{"id":"doi:10.1145/3676641.3715998","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3676641.3715998","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 30th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 2","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/A5103276362","display_name":"Y X Wang","orcid":"https://orcid.org/0009-0000-8375-493X"},"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":"Yujie Wang","raw_affiliation_strings":["Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0000-8375-493X","affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028827603","display_name":"Shiju Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I82880672","display_name":"Beihang University","ror":"https://ror.org/00wk2mp56","country_code":"CN","type":"education","lineage":["https://openalex.org/I82880672"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shiju Wang","raw_affiliation_strings":["Beihang University, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0006-5959-1652","affiliations":[{"raw_affiliation_string":"Beihang University, Beijing, China","institution_ids":["https://openalex.org/I82880672"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016753128","display_name":"Shenhan Zhu","orcid":"https://orcid.org/0009-0004-0267-775X"},"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":"Shenhan Zhu","raw_affiliation_strings":["Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0004-0267-775X","affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039254679","display_name":"Fangcheng Fu","orcid":"https://orcid.org/0000-0003-1658-0380"},"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":"Fangcheng Fu","raw_affiliation_strings":["Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-1658-0380","affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Xinyi Liu","orcid":"https://orcid.org/0009-0008-5863-2613"},"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":"Xinyi Liu","raw_affiliation_strings":["Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0008-5863-2613","affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101051966","display_name":"Xuefeng Xiao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xuefeng Xiao","raw_affiliation_strings":["ByteDance Inc., Beijing, China"],"raw_orcid":"https://orcid.org/0009-0009-8258-1243","affiliations":[{"raw_affiliation_string":"ByteDance Inc., Beijing, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":null,"display_name":"Huixia Li","orcid":"https://orcid.org/0009-0002-5968-3793"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huixia Li","raw_affiliation_strings":["ByteDance Inc., Beijing, China"],"raw_orcid":"https://orcid.org/0009-0002-5968-3793","affiliations":[{"raw_affiliation_string":"ByteDance Inc., Beijing, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014851555","display_name":"Jiashi Li","orcid":"https://orcid.org/0009-0004-7171-8527"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiashi Li","raw_affiliation_strings":["ByteDance Inc., Shenzhen, China"],"raw_orcid":"https://orcid.org/0009-0004-7171-8527","affiliations":[{"raw_affiliation_string":"ByteDance Inc., Shenzhen, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087059403","display_name":"Faming Wu","orcid":"https://orcid.org/0000-0003-4256-3277"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Faming Wu","raw_affiliation_strings":["ByteDance Inc., Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-4256-3277","affiliations":[{"raw_affiliation_string":"ByteDance Inc., Beijing, China","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5062357883","display_name":"Bin Cui","orcid":"https://orcid.org/0000-0003-1681-4677"},"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":"Bin Cui","raw_affiliation_strings":["Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-1681-4677","affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":6.7072,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.96425171,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"421","last_page":"436"},"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":0.9993000030517578,"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":0.9894000291824341,"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":0.8232733011245728},{"id":"https://openalex.org/keywords/parallelism","display_name":"Parallelism (grammar)","score":0.7383180260658264},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.6729301810264587},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6307968497276306},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.5853803157806396},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.5345242619514465},{"id":"https://openalex.org/keywords/data-parallelism","display_name":"Data parallelism","score":0.46887874603271484},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.33451876044273376}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8232733011245728},{"id":"https://openalex.org/C2781172179","wikidata":"https://www.wikidata.org/wiki/Q853109","display_name":"Parallelism (grammar)","level":2,"score":0.7383180260658264},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.6729301810264587},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6307968497276306},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.5853803157806396},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.5345242619514465},{"id":"https://openalex.org/C61483411","wikidata":"https://www.wikidata.org/wiki/Q3124522","display_name":"Data parallelism","level":3,"score":0.46887874603271484},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.33451876044273376},{"id":"https://openalex.org/C54355233","wikidata":"https://www.wikidata.org/wiki/Q7162","display_name":"Genetics","level":1,"score":0.0},{"id":"https://openalex.org/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3676641.3715998","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3676641.3715998","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 30th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 2","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":27,"referenced_works":["https://openalex.org/W2059300917","https://openalex.org/W2734941459","https://openalex.org/W2969388332","https://openalex.org/W3016430712","https://openalex.org/W3081168214","https://openalex.org/W3086105743","https://openalex.org/W3129831491","https://openalex.org/W3187255235","https://openalex.org/W3199824684","https://openalex.org/W4229005866","https://openalex.org/W4280611847","https://openalex.org/W4287235399","https://openalex.org/W4310282800","https://openalex.org/W4322718191","https://openalex.org/W4384648639","https://openalex.org/W4384918448","https://openalex.org/W4385571616","https://openalex.org/W4386768656","https://openalex.org/W4387113741","https://openalex.org/W4387356039","https://openalex.org/W4388787547","https://openalex.org/W4391032878","https://openalex.org/W4392207935","https://openalex.org/W4400893425","https://openalex.org/W4401812232","https://openalex.org/W4408362299","https://openalex.org/W6600283593"],"related_works":["https://openalex.org/W2950520577","https://openalex.org/W1501159154","https://openalex.org/W1554644772","https://openalex.org/W2003935582","https://openalex.org/W2494130044","https://openalex.org/W3170887803","https://openalex.org/W74409296","https://openalex.org/W3209384898","https://openalex.org/W4230999561","https://openalex.org/W1595672120"],"abstract_inverted_index":{"Extending":[0],"the":[1,5,44,79],"context":[2,19],"length":[3],"(i.e.,":[4,56],"maximum":[6],"supported":[7],"sequence":[8,23,35,49,54,80],"length)":[9,63],"of":[10,13,21],"LLMs":[11],"is":[12],"paramount":[14],"significance.":[15],"To":[16],"facilitate":[17],"long":[18],"training":[20,84],"LLMs,":[22],"parallelism":[24,50],"has":[25],"emerged":[26],"as":[27],"an":[28],"essential":[29],"technique,":[30],"which":[31,94],"scatters":[32],"each":[33],"input":[34,58,74],"across":[36],"multiple":[37],"devices":[38],"and":[39,64],"necessitates":[40],"communication":[41],"to":[42,96],"process":[43],"sequence.":[45],"In":[46],"essence,":[47],"existing":[48],"methods":[51],"assume":[52],"homogeneous":[53],"lengths":[55,81],"all":[57,73],"sequences":[59],"are":[60],"equal":[61],"in":[62,77,82],"therefore":[65],"leverages":[66],"a":[67,91],"single,":[68],"static":[69],"scattering":[70],"strategy":[71],"for":[72],"sequences.":[75],"However,":[76],"reality,":[78],"LLM":[83],"corpora":[85],"exhibit":[86],"substantial":[87],"variability,":[88],"often":[89],"following":[90],"long-tail":[92],"distribution,":[93],"leads":[95],"workload":[97],"heterogeneity.":[98]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":2}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}