{"id":"https://openalex.org/W7133307463","doi":"https://doi.org/10.48550/arxiv.2603.00675","title":"Specializing Foundation Models via Mixture of Low-Rank Experts for Comprehensive Head CT Analysis","display_name":"Specializing Foundation Models via Mixture of Low-Rank Experts for Comprehensive Head CT Analysis","publication_year":2026,"publication_date":"2026-02-28","ids":{"openalex":"https://openalex.org/W7133307463","doi":"https://doi.org/10.48550/arxiv.2603.00675"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.00675","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.00675","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":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.00675","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5031310336","display_name":"Youngjin Yoo","orcid":"https://orcid.org/0000-0002-0001-0503"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Yoo, Youngjin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128006993","display_name":"Han Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Han","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128026408","display_name":"Bogdan Georgescu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Georgescu, Bogdan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127980967","display_name":"Yanbo Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Yanbo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087446167","display_name":"Sa\u0161a Grbi\u0107","orcid":"https://orcid.org/0000-0002-7564-9811"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Grbic, Sasa","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127911554","display_name":"Michael Baumgartner","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Baumgartner, Michael","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055448355","display_name":"Thomas J. Re","orcid":"https://orcid.org/0000-0003-2285-1562"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Re, Thomas J.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029200200","display_name":"Jyotipriya Das","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Das, Jyotipriya","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5093393046","display_name":"Poikavila Ullaskrishnan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ullaskrishnan, Poikavila","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048357114","display_name":"Eva Eibenberger","orcid":"https://orcid.org/0000-0003-3560-830X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Eibenberger, Eva","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5019635129","display_name":"Andrei Chekkoury","orcid":"https://orcid.org/0000-0003-3651-6298"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chekkoury, Andrei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008834206","display_name":"Uttam K. Bodanapally","orcid":"https://orcid.org/0000-0003-0455-816X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bodanapally, Uttam K.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5104338801","display_name":"Savvas Nicolaou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nicolaou, Savvas","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5123835404","display_name":"Pina Sanelli","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sanelli, Pina C.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5065739324","display_name":"Thomas J. Schroeppel","orcid":"https://orcid.org/0000-0001-9164-5004"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Schroeppel, Thomas J.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5123877749","display_name":"Yvonne W. Lui","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lui, Yvonne W.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5128002978","display_name":"Eli Gibson","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gibson, Eli","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":17,"corresponding_author_ids":["https://openalex.org/A5031310336"],"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/T11775","display_name":"COVID-19 diagnosis using AI","score":0.13600000739097595,"subfield":{"id":"https://openalex.org/subfields/2741","display_name":"Radiology, Nuclear Medicine and Imaging"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},"topics":[{"id":"https://openalex.org/T11775","display_name":"COVID-19 diagnosis using AI","score":0.13600000739097595,"subfield":{"id":"https://openalex.org/subfields/2741","display_name":"Radiology, Nuclear Medicine and Imaging"},"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.10790000110864639,"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/T12422","display_name":"Radiomics and Machine Learning in Medical Imaging","score":0.10239999741315842,"subfield":{"id":"https://openalex.org/subfields/2741","display_name":"Radiology, Nuclear Medicine and Imaging"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/benchmarking","display_name":"Benchmarking","score":0.7372999787330627},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5716000199317932},{"id":"https://openalex.org/keywords/adaptation","display_name":"Adaptation (eye)","score":0.5048999786376953},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.4860000014305115},{"id":"https://openalex.org/keywords/medical-imaging","display_name":"Medical imaging","score":0.4803999960422516},{"id":"https://openalex.org/keywords/head","display_name":"Head (geology)","score":0.3815999925136566},{"id":"https://openalex.org/keywords/foundation","display_name":"Foundation (evidence)","score":0.3804999887943268},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.3790999948978424}],"concepts":[{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.7372999787330627},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6342999935150146},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5716000199317932},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5350000262260437},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5187000036239624},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.5048999786376953},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.4860000014305115},{"id":"https://openalex.org/C31601959","wikidata":"https://www.wikidata.org/wiki/Q931309","display_name":"Medical imaging","level":2,"score":0.4803999960422516},{"id":"https://openalex.org/C2780312720","wikidata":"https://www.wikidata.org/wiki/Q5689100","display_name":"Head (geology)","level":2,"score":0.3815999925136566},{"id":"https://openalex.org/C2780966255","wikidata":"https://www.wikidata.org/wiki/Q5474306","display_name":"Foundation (evidence)","level":2,"score":0.3804999887943268},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.3790999948978424},{"id":"https://openalex.org/C150899416","wikidata":"https://www.wikidata.org/wiki/Q1820378","display_name":"Transfer of learning","level":2,"score":0.37610000371932983},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.33970001339912415},{"id":"https://openalex.org/C115051666","wikidata":"https://www.wikidata.org/wiki/Q6522493","display_name":"Ranging","level":2,"score":0.3231000006198883},{"id":"https://openalex.org/C151201525","wikidata":"https://www.wikidata.org/wiki/Q177239","display_name":"Limit (mathematics)","level":2,"score":0.3203999996185303},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.30570000410079956},{"id":"https://openalex.org/C534262118","wikidata":"https://www.wikidata.org/wiki/Q177719","display_name":"Medical diagnosis","level":2,"score":0.30250000953674316},{"id":"https://openalex.org/C8038995","wikidata":"https://www.wikidata.org/wiki/Q1152135","display_name":"Unsupervised learning","level":2,"score":0.2849999964237213},{"id":"https://openalex.org/C61224824","wikidata":"https://www.wikidata.org/wiki/Q2260434","display_name":"Mixture model","level":2,"score":0.27480000257492065},{"id":"https://openalex.org/C82157600","wikidata":"https://www.wikidata.org/wiki/Q2671652","display_name":"Diagnostic test","level":2,"score":0.25200000405311584}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.00675","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.00675","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":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.00675","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.00675","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":false,"raw_source_name":null,"raw_type":"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":{"Foundation":[0],"models":[1,98,154,169],"pre-trained":[2],"on":[3,197],"large-scale":[4],"datasets":[5],"demonstrate":[6,141],"strong":[7],"transfer":[8],"learning":[9],"capabilities;":[10],"however,":[11],"their":[12],"adaptation":[13,73],"to":[14,116],"complex":[15],"multi-label":[16],"diagnostic":[17],"tasks-such":[18],"as":[19,31,201],"comprehensive":[20,88],"head":[21,107,123],"CT":[22,124],"finding":[23],"detection-remains":[24],"understudied.":[25],"Standard":[26],"parameter-efficient":[27],"fine-tuning":[28],"methods":[29],"such":[30],"LoRA":[32,58],"apply":[33],"uniform":[34],"adaptations":[35],"across":[36,92,145],"pathology":[37,83],"types,":[38],"which":[39],"may":[40],"limit":[41],"performance":[42,143],"for":[43],"diverse":[44],"medical":[45,95],"findings.":[46],"We":[47,85],"propose":[48],"a":[49,87],"Mixture":[50],"of":[51,90,177,187,194],"Low-Rank":[52],"Experts":[53],"(MoLRE)":[54],"framework":[55],"that":[56],"extends":[57],"with":[59,74,126],"multiple":[60],"specialized":[61],"low-rank":[62],"adapters":[63],"and":[64,80,101,106,110,137,152,179,205],"unsupervised":[65],"soft":[66],"routing.":[67],"This":[68],"approach":[69],"enables":[70],"conditional":[71],"feature":[72],"less":[75],"than":[76],"0.5%":[77],"additional":[78],"parameters":[79],"without":[81],"explicit":[82],"supervision.":[84],"present":[86],"benchmark":[89],"MoLRE":[91,178],"six":[93],"state-of-the-art":[94],"imaging":[96],"foundation":[97],"spanning":[99],"2D":[100],"3D":[102,164],"architectures,":[103],"general-domain,":[104],"medical-domain,":[105],"CT-specific":[108],"pretraining,":[109],"model":[111,206],"sizes":[112],"ranging":[113],"from":[114],"7M":[115],"431M":[117],"parameters.":[118],"Using":[119],"over":[120],"70,000":[121],"non-contrast":[122],"scans":[125],"75":[127],"annotated":[128],"findings-including":[129],"hemorrhage,":[130],"infarction,":[131],"trauma,":[132],"mass":[133],"lesions,":[134],"structural":[135],"abnormalities,":[136],"chronic":[138],"changes-our":[139],"experiments":[140],"consistent":[142],"improvements":[144,158],"all":[146],"models.":[147],"Gains":[148],"vary":[149],"substantially:":[150],"general-purpose":[151],"medical-domain":[153],"show":[155,170],"the":[156,182,192],"largest":[157],"(DINOv3-Base:":[159],"+4.6%;":[160],"MedGemma:":[161],"+4.3%),":[162],"whereas":[163],"CT-specialized":[165],"or":[166],"very":[167],"large":[168],"more":[171],"modest":[172],"gains":[173],"(+0.2-1.3%).":[174],"The":[175],"combination":[176],"MedGemma":[180],"achieves":[181],"highest":[183],"average":[184],"detection":[185],"AUC":[186],"0.917.":[188],"These":[189],"findings":[190],"highlight":[191],"importance":[193],"systematic":[195],"benchmarking":[196],"target":[198],"clinical":[199],"tasks,":[200],"pretraining":[202],"domain,":[203],"architecture,":[204],"scale":[207],"interact":[208],"in":[209],"non-obvious":[210],"ways.":[211]},"counts_by_year":[],"updated_date":"2026-05-04T08:30:34.212998","created_date":"2026-03-04T00:00:00"}