{"id":"https://openalex.org/W7160957168","doi":"https://doi.org/10.48550/arxiv.2605.10840","title":"Clin-JEPA: A Multi-Phase Co-Training Framework for Joint-Embedding Predictive Pretraining on EHR Patient Trajectories","display_name":"Clin-JEPA: A Multi-Phase Co-Training Framework for Joint-Embedding Predictive Pretraining on EHR Patient Trajectories","publication_year":2026,"publication_date":"2026-05-11","ids":{"openalex":"https://openalex.org/W7160957168","doi":"https://doi.org/10.48550/arxiv.2605.10840"},"language":"en","primary_location":{"id":"pmh:oai:pubmedcentral.nih.gov:13193276","is_oa":true,"landing_page_url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC13193276/","pdf_url":null,"source":{"id":"https://openalex.org/S2764455111","display_name":"PubMed Central","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"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":"ArXiv","raw_type":"Text"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC13193276/","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5135980830","display_name":"Yixuan Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Yixuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062046627","display_name":"Mehak Arora","orcid":"https://orcid.org/0000-0003-1225-9549"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Arora, Mehak","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054393648","display_name":"Ryan Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Ryan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119381014","display_name":"Baraa Abed","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Abed, Baraa","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5011517497","display_name":"Junseob Kim","orcid":"https://orcid.org/0000-0002-6959-5271"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kim, Junseob","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5037399650","display_name":"Tilendra Choudhary","orcid":"https://orcid.org/0000-0001-5012-9046"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Choudhary, Tilendra","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136001046","display_name":"Md Hassanuzzaman","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hassanuzzaman, Md","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135909896","display_name":"Kevin Zhu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhu, Kevin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135997630","display_name":"Ayman Ali","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ali, Ayman","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135930141","display_name":"Chengkun Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Chengkun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5019495781","display_name":"A.E. Gent","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gent, Alasdair Edward","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5049082850","display_name":"Victor Moas","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Moas, Victor","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5012511062","display_name":"Rishikesan Kamaleswaran","orcid":"https://orcid.org/0000-0001-8366-4811"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kamaleswaran, Rishikesan","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/T13702","display_name":"Machine Learning in Healthcare","score":0.2874999940395355,"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/T13702","display_name":"Machine Learning in Healthcare","score":0.2874999940395355,"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/T13248","display_name":"Healthcare Technology and Patient Monitoring","score":0.11379999667406082,"subfield":{"id":"https://openalex.org/subfields/2746","display_name":"Surgery"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.066600002348423,"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/encoder","display_name":"Encoder","score":0.7386999726295471},{"id":"https://openalex.org/keywords/discriminative-model","display_name":"Discriminative model","score":0.6172000169754028},{"id":"https://openalex.org/keywords/baseline","display_name":"Baseline (sea)","score":0.5989999771118164},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.49950000643730164},{"id":"https://openalex.org/keywords/binary-number","display_name":"Binary number","score":0.4970000088214874},{"id":"https://openalex.org/keywords/trajectory","display_name":"Trajectory","score":0.4823000133037567},{"id":"https://openalex.org/keywords/autoregressive-model","display_name":"Autoregressive model","score":0.4717999994754791},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.4684999883174896},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.42660000920295715}],"concepts":[{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.7386999726295471},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6291000247001648},{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.6172000169754028},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.5989999771118164},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5724999904632568},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.49950000643730164},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.4970000088214874},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4950000047683716},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.4823000133037567},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.4717999994754791},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.4684999883174896},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.42660000920295715},{"id":"https://openalex.org/C66905080","wikidata":"https://www.wikidata.org/wiki/Q17005494","display_name":"Binary classification","level":3,"score":0.3732999861240387},{"id":"https://openalex.org/C2776207758","wikidata":"https://www.wikidata.org/wiki/Q5303302","display_name":"Downstream (manufacturing)","level":2,"score":0.36730000376701355},{"id":"https://openalex.org/C34413123","wikidata":"https://www.wikidata.org/wiki/Q170978","display_name":"Robotics","level":3,"score":0.3553999960422516},{"id":"https://openalex.org/C2779843651","wikidata":"https://www.wikidata.org/wiki/Q7390335","display_name":"SIGNAL (programming language)","level":2,"score":0.3386000096797943},{"id":"https://openalex.org/C45804977","wikidata":"https://www.wikidata.org/wiki/Q7239673","display_name":"Predictive modelling","level":2,"score":0.33169999718666077},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.31529998779296875},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.303600013256073},{"id":"https://openalex.org/C101738243","wikidata":"https://www.wikidata.org/wiki/Q786435","display_name":"Autoencoder","level":3,"score":0.3012999892234802},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.2825999855995178},{"id":"https://openalex.org/C51167844","wikidata":"https://www.wikidata.org/wiki/Q4422623","display_name":"Latent variable","level":2,"score":0.27320000529289246},{"id":"https://openalex.org/C48677424","wikidata":"https://www.wikidata.org/wiki/Q6888088","display_name":"Mode (computer interface)","level":2,"score":0.26919999718666077},{"id":"https://openalex.org/C66746571","wikidata":"https://www.wikidata.org/wiki/Q1134833","display_name":"ENCODE","level":3,"score":0.2621000111103058},{"id":"https://openalex.org/C2779803651","wikidata":"https://www.wikidata.org/wiki/Q5282088","display_name":"Discriminator","level":3,"score":0.2522999942302704}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:pubmedcentral.nih.gov:13193276","is_oa":true,"landing_page_url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC13193276/","pdf_url":null,"source":{"id":"https://openalex.org/S2764455111","display_name":"PubMed Central","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"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":"ArXiv","raw_type":"Text"},{"id":"doi:10.48550/arxiv.2605.10840","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.10840","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":"pmh:oai:pubmedcentral.nih.gov:13193276","is_oa":true,"landing_page_url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC13193276/","pdf_url":null,"source":{"id":"https://openalex.org/S2764455111","display_name":"PubMed Central","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"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":"ArXiv","raw_type":"Text"},"sustainable_development_goals":[{"score":0.7269149422645569,"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"We":[0],"present":[1],"Clin-JEPA,":[2],"a":[3,40,77,105,156,160,198,220],"multi-phase":[4],"co-training":[5,99,116,155],"framework":[6],"for":[7,216],"joint-embedding":[8],"predictive":[9],"(JEPA)":[10],"pretraining":[11,70,132],"on":[12,76,229,238,243],"EHR":[13,35],"patient":[14,46],"trajectories.":[15],"JEPA":[16,63,107],"architectures":[17],"have":[18],"enabled":[19],"latent-space":[20],"planning":[21],"in":[22,28,211],"robotics":[23],"and":[24,48,102,122,144,159,188,226,241,249],"high-quality":[25],"representation":[26,120],"learning":[27],"vision,":[29],"but":[30,114],"extending":[31],"the":[32,67,84,88,92,100,173,195],"paradigm":[33],"to":[34,38,128,192],"data":[36],"--":[37,57,134,147],"obtain":[39],"single":[41,221],"backbone":[42,222],"that":[43,91],"simultaneously":[44],"forecasts":[45],"trajectories":[47],"serves":[49],"diverse":[50],"downstream":[51,231],"risk-prediction":[52],"tasks":[53,247],"without":[54],"per-task":[55],"fine-tuning":[56],"remains":[58],"an":[59],"open":[60],"challenge.":[61],"Existing":[62],"frameworks":[64],"either":[65],"discard":[66],"predictor":[68,94,103,135,145],"after":[69],"(I-JEPA,":[71],"V-JEPA)":[72],"or":[73],"train":[74],"it":[75],"frozen":[78],"pretrained":[79],"encoder":[80,85,101,158,196],"(V-JEPA":[81],"2-AC),":[82],"leaving":[83],"unaware":[86],"of":[87],"rollout":[89,127,178],"signal":[90],"retained":[93],"must":[95],"use":[96],"at":[97],"inference;":[98],"under":[104],"shared":[106],"prediction":[108],"objective":[109],"would":[110],"supply":[111],"this":[112],"grounding,":[113],"na\u00efve":[115],"is":[117],"unstable,":[118],"with":[119],"collapse":[121],"online/target":[123],"drift":[124,179],"causing":[125],"autoregressive":[126],"diverge.":[129],"Clin-JEPA's":[130],"five-phase":[131],"curriculum":[133],"warmup,":[136],"joint":[137],"refinement,":[138],"EMA":[139],"target":[140],"alignment,":[141],"hard":[142],"sync,":[143],"finalization":[146],"addresses":[148],"each":[149],"failure":[150],"mode":[151],"by":[152],"phase,":[153],"stably":[154],"Qwen3-8B-based":[157],"92M-parameter":[161],"latent":[162,176,201,212],"trajectory":[163],"predictor.":[164],"On":[165],"MIMIC-IV":[166],"ICU":[167],"data,":[168],"three":[169],"independent":[170],"evaluations":[171],"support":[172],"framework:":[174],"(1)":[175],"$\\ell_1$":[177],"uniquely":[180],"converges":[181],"($-$15.7%)":[182],"over":[183],"48-hour":[184],"horizons":[185],"while":[186],"baselines":[187,228],"ablations":[189],"diverge":[190],"(+3%":[191],"+4951%);":[193],"(2)":[194],"learns":[197],"clinically":[199],"discriminative":[200],"geometry":[202],"(deteriorating-patient":[203],"cohorts":[204],"displace":[205],"4.83$\\times$":[206],"further":[207],"than":[208],"stable":[209],"patients":[210],"space,":[213],"vs":[214,251],"$\\leq$2.62$\\times$":[215],"baseline":[217,252],"encoders);":[218],"(3)":[219],"outperforms":[223],"strong":[224],"tabular":[225],"sequence":[227],"multi-task":[230],"evaluation.":[232],"Clin-JEPA":[233],"achieves":[234],"mean":[235],"AUROC":[236],"0.851":[237],"ICareFM":[239],"EEP":[240],"0.883":[242],"8":[244],"binary":[245],"risk":[246],"(+0.038":[248],"+0.041":[250],"average).":[253]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-05-13T00:00:00"}