{"id":"https://openalex.org/W4416199186","doi":"https://doi.org/10.1145/3712285.3759855","title":"SlimPipe: Memory-Thrifty and Efficient Pipeline Parallelism for Long-Context LLM Training","display_name":"SlimPipe: Memory-Thrifty and Efficient Pipeline Parallelism for Long-Context LLM Training","publication_year":2025,"publication_date":"2025-11-12","ids":{"openalex":"https://openalex.org/W4416199186","doi":"https://doi.org/10.1145/3712285.3759855"},"language":null,"primary_location":{"id":"doi:10.1145/3712285.3759855","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3712285.3759855","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis","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/A5036757729","display_name":"Zhouyang Li","orcid":"https://orcid.org/0009-0004-7317-1166"},"institutions":[{"id":"https://openalex.org/I4401726859","display_name":"Kuaishou (China)","ror":"https://ror.org/0258as409","country_code":null,"type":"company","lineage":["https://openalex.org/I4401726859"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Zhouyang Li","raw_affiliation_strings":["Kuaishou Technology, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Kuaishou Technology, Beijing, China","institution_ids":["https://openalex.org/I4401726859"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102852473","display_name":"Yuliang Liu","orcid":"https://orcid.org/0009-0003-1878-0199"},"institutions":[{"id":"https://openalex.org/I4401726859","display_name":"Kuaishou (China)","ror":"https://ror.org/0258as409","country_code":null,"type":"company","lineage":["https://openalex.org/I4401726859"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuliang Liu","raw_affiliation_strings":["Kuaishou Technology, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Kuaishou Technology, Beijing, China","institution_ids":["https://openalex.org/I4401726859"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5092512993","display_name":"Wei Zhang","orcid":"https://orcid.org/0009-0009-5071-489X"},"institutions":[{"id":"https://openalex.org/I4401726859","display_name":"Kuaishou (China)","ror":"https://ror.org/0258as409","country_code":null,"type":"company","lineage":["https://openalex.org/I4401726859"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wei Zhang","raw_affiliation_strings":["Kuaishou Technology, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Kuaishou Technology, Beijing, China","institution_ids":["https://openalex.org/I4401726859"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087755974","display_name":"Tailing Yuan","orcid":"https://orcid.org/0000-0002-6119-8829"},"institutions":[{"id":"https://openalex.org/I4401726859","display_name":"Kuaishou (China)","ror":"https://ror.org/0258as409","country_code":null,"type":"company","lineage":["https://openalex.org/I4401726859"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tailing Yuan","raw_affiliation_strings":["Kuaishou Technology, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Kuaishou Technology, Beijing, China","institution_ids":["https://openalex.org/I4401726859"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5092423906","display_name":"Bin Chen","orcid":"https://orcid.org/0009-0008-7953-1651"},"institutions":[{"id":"https://openalex.org/I4401726859","display_name":"Kuaishou (China)","ror":"https://ror.org/0258as409","country_code":null,"type":"company","lineage":["https://openalex.org/I4401726859"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bin Chen","raw_affiliation_strings":["Kuaishou Technology, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Kuaishou Technology, Beijing, China","institution_ids":["https://openalex.org/I4401726859"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5070531404","display_name":"Chengru Song","orcid":"https://orcid.org/0009-0009-3826-8436"},"institutions":[{"id":"https://openalex.org/I4401726859","display_name":"Kuaishou (China)","ror":"https://ror.org/0258as409","country_code":null,"type":"company","lineage":["https://openalex.org/I4401726859"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chengru Song","raw_affiliation_strings":["Kuaishou Technology, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Kuaishou Technology, Beijing, China","institution_ids":["https://openalex.org/I4401726859"]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5036757729"],"corresponding_institution_ids":["https://openalex.org/I4401726859"],"apc_list":null,"apc_paid":null,"fwci":4.9698,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.95584144,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1409","last_page":"1428"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.1800999939441681,"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":0.1800999939441681,"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":0.14090000092983246,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.0877000018954277,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/pipeline","display_name":"Pipeline (software)","score":0.8342999815940857},{"id":"https://openalex.org/keywords/parallelism","display_name":"Parallelism (grammar)","score":0.7706999778747559},{"id":"https://openalex.org/keywords/pipeline-transport","display_name":"Pipeline transport","score":0.4855000078678131},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.3926999866962433},{"id":"https://openalex.org/keywords/data-parallelism","display_name":"Data parallelism","score":0.3919000029563904},{"id":"https://openalex.org/keywords/energy-consumption","display_name":"Energy consumption","score":0.3176000118255615}],"concepts":[{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.8342999815940857},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7807999849319458},{"id":"https://openalex.org/C2781172179","wikidata":"https://www.wikidata.org/wiki/Q853109","display_name":"Parallelism (grammar)","level":2,"score":0.7706999778747559},{"id":"https://openalex.org/C175309249","wikidata":"https://www.wikidata.org/wiki/Q725864","display_name":"Pipeline transport","level":2,"score":0.4855000078678131},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.444599986076355},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.3926999866962433},{"id":"https://openalex.org/C61483411","wikidata":"https://www.wikidata.org/wiki/Q3124522","display_name":"Data parallelism","level":3,"score":0.3919000029563904},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.3176000118255615},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.3158999979496002},{"id":"https://openalex.org/C30772137","wikidata":"https://www.wikidata.org/wiki/Q5164762","display_name":"Consumption (sociology)","level":2,"score":0.3124000132083893},{"id":"https://openalex.org/C140763907","wikidata":"https://www.wikidata.org/wiki/Q2714055","display_name":"Instruction-level parallelism","level":3,"score":0.2992999851703644},{"id":"https://openalex.org/C42992933","wikidata":"https://www.wikidata.org/wiki/Q691169","display_name":"Task parallelism","level":3,"score":0.28929999470710754},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.27639999985694885}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3712285.3759855","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3712285.3759855","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis","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":12,"referenced_works":["https://openalex.org/W1498436455","https://openalex.org/W2579247884","https://openalex.org/W2964174152","https://openalex.org/W2969388332","https://openalex.org/W3081168214","https://openalex.org/W3129831491","https://openalex.org/W3132107458","https://openalex.org/W3204998121","https://openalex.org/W4385571616","https://openalex.org/W4387321091","https://openalex.org/W4389518760","https://openalex.org/W4411120118"],"related_works":[],"abstract_inverted_index":{"Pipeline":[0],"Parallelism":[1],"serves":[2],"as":[3,12],"a":[4],"crucial":[5],"technique":[6],"for":[7],"training":[8],"Large":[9],"Language":[10],"Models,":[11],"it":[13],"alleviates":[14],"memory":[15,39,46],"pressure":[16],"from":[17,49],"model":[18],"states":[19],"with":[20],"relatively":[21],"low":[22],"communication":[23],"overhead.":[24],"However,":[25],"in":[26],"long-context":[27],"scenarios,":[28],"existing":[29],"pipeline":[30,63],"parallelism":[31],"methods":[32],"fail":[33],"to":[34,43],"address":[35],"the":[36,44,50],"substantial":[37],"activation":[38],"pressure,":[40],"primarily":[41],"due":[42],"peak":[45],"consumption":[47],"resulting":[48],"accumulation":[51],"of":[52],"activations":[53],"across":[54],"multiple":[55],"microbatches.":[56],"Moreover,":[57],"these":[58],"approaches":[59],"inevitably":[60],"introduce":[61],"considerable":[62],"bubbles,":[64],"further":[65],"hindering":[66],"efficiency.":[67]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1}],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-11-12T00:00:00"}