{"id":"https://openalex.org/W4415333939","doi":"https://doi.org/10.1145/3805621.3807638","title":"All is Not Lost: LLM Recovery without Checkpoints","display_name":"All is Not Lost: LLM Recovery without Checkpoints","publication_year":2026,"publication_date":"2026-04-27","ids":{"openalex":"https://openalex.org/W4415333939","doi":"https://doi.org/10.1145/3805621.3807638"},"language":"en","primary_location":{"id":"doi:10.1145/3805621.3807638","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805621.3807638","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Sixth European Workshop on Machine Learning and Systems","raw_type":"proceedings-article"},"type":"article","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3805621.3807638","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5120057248","display_name":"Nikolay Blagoev","orcid":null},"institutions":[{"id":"https://openalex.org/I57825437","display_name":"University of Neuch\u00e2tel","ror":"https://ror.org/00vasag41","country_code":"CH","type":"education","lineage":["https://openalex.org/I57825437"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Nikolay Blagoev","raw_affiliation_strings":["Gensyn, West Hollywood, USA and Computer Science Department, University of Neuch\u00e2tel, Neuch\u00e2tel, Switzerland"],"raw_orcid":"https://orcid.org/0009-0006-1364-0391","affiliations":[{"raw_affiliation_string":"Gensyn, West Hollywood, USA and Computer Science Department, University of Neuch\u00e2tel, Neuch\u00e2tel, Switzerland","institution_ids":["https://openalex.org/I57825437"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120057249","display_name":"O\u011fuzhan Ersoy","orcid":null},"institutions":[{"id":"https://openalex.org/I4210134910","display_name":"Sheridan Technical College","ror":"https://ror.org/03693et41","country_code":"US","type":"education","lineage":["https://openalex.org/I4210134910"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"O\u011fuzhan Ersoy","raw_affiliation_strings":["Gensyn, WEST HOLLYWOOD, USA"],"raw_orcid":"https://orcid.org/0000-0002-6428-4212","affiliations":[{"raw_affiliation_string":"Gensyn, WEST HOLLYWOOD, USA","institution_ids":["https://openalex.org/I4210134910"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5013859152","display_name":"Lydia Y. Chen","orcid":"https://orcid.org/0000-0002-4228-6735"},"institutions":[{"id":"https://openalex.org/I57825437","display_name":"University of Neuch\u00e2tel","ror":"https://ror.org/00vasag41","country_code":"CH","type":"education","lineage":["https://openalex.org/I57825437"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Lydia Yiyu Chen","raw_affiliation_strings":["Computer Science Department, University of Neuch\u00e2tel, Neuch\u00e2tel, Switzerland"],"raw_orcid":"https://orcid.org/0000-0002-4228-6735","affiliations":[{"raw_affiliation_string":"Computer Science Department, University of Neuch\u00e2tel, Neuch\u00e2tel, Switzerland","institution_ids":["https://openalex.org/I57825437"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.00150091,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"323","last_page":"332"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12509","display_name":"Corporate Insolvency and Governance","score":0.7587000131607056,"subfield":{"id":"https://openalex.org/subfields/1402","display_name":"Accounting"},"field":{"id":"https://openalex.org/fields/14","display_name":"Business, Management and Accounting"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T12509","display_name":"Corporate Insolvency and Governance","score":0.7587000131607056,"subfield":{"id":"https://openalex.org/subfields/1402","display_name":"Accounting"},"field":{"id":"https://openalex.org/fields/14","display_name":"Business, Management and Accounting"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T11762","display_name":"Law, Economics, and Judicial Systems","score":0.6851000189781189,"subfield":{"id":"https://openalex.org/subfields/2002","display_name":"Economics and Econometrics"},"field":{"id":"https://openalex.org/fields/20","display_name":"Economics, Econometrics and Finance"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.7222999930381775},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.6122000217437744},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.5946999788284302},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.5927000045776367},{"id":"https://openalex.org/keywords/boosting","display_name":"Boosting (machine learning)","score":0.4131999909877777},{"id":"https://openalex.org/keywords/copying","display_name":"Copying","score":0.33739998936653137}],"concepts":[{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.7222999930381775},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6960999965667725},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.6122000217437744},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.5946999788284302},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.5927000045776367},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.4880000054836273},{"id":"https://openalex.org/C46686674","wikidata":"https://www.wikidata.org/wiki/Q466303","display_name":"Boosting (machine learning)","level":2,"score":0.4131999909877777},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.3589000105857849},{"id":"https://openalex.org/C200601418","wikidata":"https://www.wikidata.org/wiki/Q2193887","display_name":"Reliability engineering","level":1,"score":0.3434000015258789},{"id":"https://openalex.org/C2779151265","wikidata":"https://www.wikidata.org/wiki/Q1156791","display_name":"Copying","level":2,"score":0.33739998936653137},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.32429999113082886},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.3174999952316284},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.2858999967575073},{"id":"https://openalex.org/C63540848","wikidata":"https://www.wikidata.org/wiki/Q3140932","display_name":"Fault tolerance","level":2,"score":0.2547999918460846}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1145/3805621.3807638","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805621.3807638","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Sixth European Workshop on Machine Learning and Systems","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2506.15461","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2506.15461","pdf_url":"https://arxiv.org/pdf/2506.15461","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2506.15461","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2506.15461","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.1145/3805621.3807638","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805621.3807638","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Sixth European Workshop on Machine Learning and Systems","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Training":[0],"Large":[1],"Language":[2],"Models":[3],"on":[4],"decentralized":[5],"nodes,":[6],"e.g.,":[7],"on-spot":[8],"instances,":[9],"lowers":[10],"the":[11,23,32,41,62],"training":[12],"cost":[13],"and":[14,31,86],"enables":[15],"model":[16,42,64],"democratization.":[17],"The":[18,45],"inevitable":[19],"challenge":[20],"here":[21],"is":[22,52,65],"transient":[24],"churns":[25],"of":[26,40,61],"nodes":[27],"due":[28],"to":[29,37,48,53,67],"failures":[30,51],"operator's":[33],"scheduling":[34],"policies,":[35],"leading":[36],"losing":[38],"parts":[39],"(some":[43],"layers).":[44],"conventional":[46],"approaches":[47,75],"recover":[49],"from":[50],"either":[54],"use":[55],"checkpointing,":[56],"where":[57],"periodically":[58],"a":[59],"copy":[60],"entire":[63],"sent":[66],"an":[68],"additional":[69],"storage,":[70],"or":[71],"redundant":[72],"computation.":[73],"These":[74],"yield":[76],"significant":[77],"communication":[78],"and/or":[79],"computation":[80],"overhead":[81],"even":[82],"in":[83,89],"non-failure":[84],"cases":[85],"scale":[87],"poorly":[88],"settings":[90],"with":[91],"large":[92],"models.":[93]},"counts_by_year":[],"updated_date":"2026-07-03T08:13:44.112507","created_date":"2025-10-19T00:00:00"}