{"id":"https://openalex.org/W7154453544","doi":"https://doi.org/10.48550/arxiv.2604.12391","title":"Chain-of-Models Pre-Training: Rethinking Training Acceleration of Vision Foundation Models","display_name":"Chain-of-Models Pre-Training: Rethinking Training Acceleration of Vision Foundation Models","publication_year":2026,"publication_date":"2026-04-14","ids":{"openalex":"https://openalex.org/W7154453544","doi":"https://doi.org/10.48550/arxiv.2604.12391"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.12391","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.12391","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2604.12391","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5133679545","display_name":"Jiawei Fan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fan, Jiawei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5013236807","display_name":"Shigeng Wang","orcid":"https://orcid.org/0009-0006-6878-9652"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Shigeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100388742","display_name":"Chao Li","orcid":"https://orcid.org/0000-0003-2772-3244"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Chao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100393835","display_name":"Xiaolong Liu","orcid":"https://orcid.org/0009-0000-3427-9255"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Xiaolong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5025017745","display_name":"Anbang Yao","orcid":"https://orcid.org/0000-0002-3878-8679"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yao, Anbang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.5146999955177307,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.5146999955177307,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.2556000053882599,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T10036","display_name":"Advanced Neural Network Applications","score":0.03620000183582306,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/acceleration","display_name":"Acceleration","score":0.7386999726295471},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.5534999966621399},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.5533000230789185},{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.5461000204086304},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.474700003862381},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.45820000767707825},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.44859999418258667},{"id":"https://openalex.org/keywords/frame","display_name":"Frame (networking)","score":0.4050999879837036},{"id":"https://openalex.org/keywords/property","display_name":"Property (philosophy)","score":0.3617999851703644}],"concepts":[{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.7386999726295471},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6549000144004822},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.5534999966621399},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.5533000230789185},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.5461000204086304},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5094000101089478},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.474700003862381},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.45820000767707825},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.44859999418258667},{"id":"https://openalex.org/C126042441","wikidata":"https://www.wikidata.org/wiki/Q1324888","display_name":"Frame (networking)","level":2,"score":0.4050999879837036},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4027999937534332},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3853999972343445},{"id":"https://openalex.org/C189950617","wikidata":"https://www.wikidata.org/wiki/Q937228","display_name":"Property (philosophy)","level":2,"score":0.3617999851703644},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.3513999879360199},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.3208000063896179},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.31360000371932983},{"id":"https://openalex.org/C2164484","wikidata":"https://www.wikidata.org/wiki/Q5170150","display_name":"Core (optical fiber)","level":2,"score":0.31049999594688416},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3089999854564667},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.3066999912261963},{"id":"https://openalex.org/C207467116","wikidata":"https://www.wikidata.org/wiki/Q4385666","display_name":"Inverse","level":2,"score":0.303600013256073},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.28130000829696655},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.2759999930858612},{"id":"https://openalex.org/C104114177","wikidata":"https://www.wikidata.org/wiki/Q79782","display_name":"Motion (physics)","level":2,"score":0.27459999918937683},{"id":"https://openalex.org/C182306322","wikidata":"https://www.wikidata.org/wiki/Q1779371","display_name":"Order (exchange)","level":2,"score":0.2630999982357025},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.258899986743927},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.25699999928474426},{"id":"https://openalex.org/C114289077","wikidata":"https://www.wikidata.org/wiki/Q3284399","display_name":"Statistical model","level":2,"score":0.25200000405311584}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.12391","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.12391","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.12391","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.12391","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"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":{"In":[0,77],"this":[1,78,141],"paper,":[2],"we":[3,240],"present":[4],"Chain-of-Models":[5],"Pre-Training":[6],"(CoM-PT),":[7],"a":[8,60,118,159,200,222],"novel":[9],"performance-lossless":[10],"training":[11,43,134,138,162],"acceleration":[12,25,217],"method":[13],"for":[14,63],"vision":[15],"foundation":[16],"models":[17,91,123,164,186],"(VFMs).":[18],"This":[19],"approach":[20],"fundamentally":[21],"differs":[22],"from":[23,100,225],"existing":[24],"methods":[26],"in":[27,68,109,167,248],"its":[28,154],"core":[29],"motivation:":[30],"rather":[31],"than":[32],"optimizing":[33],"each":[34],"model":[35,47,54,65,72,75,83,189,202,257],"individually,":[36],"CoM-PT":[37,58,120,220,232],"is":[38,128,142,233],"designed":[39],"to":[40,124,131,187,196,227,236,244],"accelerate":[41],"the":[42,46,53,64,81,89,107,110,114,180,188,206,216,242],"pipeline":[44],"at":[45],"family":[48,55,208],"level,":[49],"scaling":[50,156],"efficiently":[51,93],"as":[52,179,205,254],"expands.":[56],"Specifically,":[57],"establishes":[59],"pre-training":[61,173,238],"sequence":[62],"family,":[66],"arranged":[67],"ascending":[69],"order":[70],"of":[71,219],"size,":[73],"called":[74],"chain.":[76],"chain,":[79],"only":[80],"smallest":[82],"undergoes":[84],"standard":[85,132],"individual":[86,133],"pre-training,":[87],"while":[88,135],"other":[90],"are":[92],"trained":[94],"through":[95],"sequential":[96],"inverse":[97],"knowledge":[98,108],"transfer":[99],"their":[101],"smaller":[102,185],"predecessors":[103],"by":[104,194],"jointly":[105],"reusing":[106],"parameter":[111],"space":[112],"and":[113,140,150,213,229],"feature":[115],"space.":[116],"As":[117],"result,":[119],"enables":[121],"all":[122],"achieve":[125],"performance":[126],"that":[127],"mostly":[129],"superior":[130],"significantly":[136],"reducing":[137],"cost,":[139],"extensively":[143],"validated":[144],"across":[145,210],"45":[146],"datasets":[147],"spanning":[148],"zero-shot":[149],"fine-tuning":[151],"tasks.":[152],"Notably,":[153],"efficient":[155],"property":[157],"yields":[158],"remarkable":[160],"phenomenon:":[161],"more":[163,249],"even":[165],"results":[166],"higher":[168],"efficiency.":[169],"For":[170],"instance,":[171],"when":[172],"on":[174],"CC3M:":[175],"i)":[176],"given":[177],"ViT-L":[178],"largest":[181],"model,":[182],"progressively":[183],"prepending":[184],"chain":[190],"reduces":[191],"computational":[192],"complexity":[193],"up":[195],"72%;":[197],"ii)":[198],"within":[199],"fixed":[201],"size":[203],"range,":[204],"VFM":[207],"scales":[209],"3,":[211],"4,":[212],"7":[214],"models,":[215],"ratio":[218],"exhibits":[221],"striking":[223],"leap:":[224],"4.13X":[226],"5.68X":[228],"7.09X.":[230],"Since":[231],"naturally":[234],"agnostic":[235],"specific":[237],"paradigms,":[239],"open-source":[241],"code":[243],"spur":[245],"further":[246],"extensions":[247],"computationally":[250],"intensive":[251],"scenarios,":[252],"such":[253],"large":[255],"language":[256],"pre-training.":[258]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-04-16T00:00:00"}