{"id":"https://openalex.org/W7152991653","doi":"https://doi.org/10.48550/arxiv.2604.07779","title":"Plug-and-Play Logit Fusion for Heterogeneous Pathology Foundation Models","display_name":"Plug-and-Play Logit Fusion for Heterogeneous Pathology Foundation Models","publication_year":2026,"publication_date":"2026-04-09","ids":{"openalex":"https://openalex.org/W7152991653","doi":"https://doi.org/10.48550/arxiv.2604.07779"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.07779","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.07779","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":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2604.07779","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5079788591","display_name":"Gexin Huang","orcid":"https://orcid.org/0000-0003-1390-9988"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Huang, Gexin","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133350429","display_name":"Anqi Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Anqi","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133376918","display_name":"Yusheng Tan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tan, Yusheng","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5072638366","display_name":"Beidi Zhao","orcid":"https://orcid.org/0000-0002-2845-7550"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Beidi","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133331601","display_name":"Gang Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Gang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Gao, Zu-Hua","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gao, Zu-Hua","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5133348039","display_name":"Xiaoxiao Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Xiaoxiao","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5079788591"],"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/T10862","display_name":"AI in cancer detection","score":0.9801999926567078,"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/T10862","display_name":"AI in cancer detection","score":0.9801999926567078,"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/T12422","display_name":"Radiomics and Machine Learning in Medical Imaging","score":0.002300000051036477,"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/T12874","display_name":"Digital Imaging for Blood Diseases","score":0.0019000000320374966,"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/benchmark","display_name":"Benchmark (surveying)","score":0.414900004863739},{"id":"https://openalex.org/keywords/retraining","display_name":"Retraining","score":0.3946000039577484},{"id":"https://openalex.org/keywords/unobservable","display_name":"Unobservable","score":0.3889000117778778},{"id":"https://openalex.org/keywords/range","display_name":"Range (aeronautics)","score":0.3824999928474426},{"id":"https://openalex.org/keywords/fusion","display_name":"Fusion","score":0.37549999356269836},{"id":"https://openalex.org/keywords/foundation","display_name":"Foundation (evidence)","score":0.3659999966621399},{"id":"https://openalex.org/keywords/logit","display_name":"Logit","score":0.33640000224113464},{"id":"https://openalex.org/keywords/product","display_name":"Product (mathematics)","score":0.3244999945163727}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6912000179290771},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5091000199317932},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4645000100135803},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.414900004863739},{"id":"https://openalex.org/C2778712577","wikidata":"https://www.wikidata.org/wiki/Q3505966","display_name":"Retraining","level":2,"score":0.3946000039577484},{"id":"https://openalex.org/C2780695315","wikidata":"https://www.wikidata.org/wiki/Q3799040","display_name":"Unobservable","level":2,"score":0.3889000117778778},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.3824999928474426},{"id":"https://openalex.org/C158525013","wikidata":"https://www.wikidata.org/wiki/Q2593739","display_name":"Fusion","level":2,"score":0.37549999356269836},{"id":"https://openalex.org/C2780966255","wikidata":"https://www.wikidata.org/wiki/Q5474306","display_name":"Foundation (evidence)","level":2,"score":0.3659999966621399},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3472000062465668},{"id":"https://openalex.org/C140331021","wikidata":"https://www.wikidata.org/wiki/Q1868104","display_name":"Logit","level":2,"score":0.33640000224113464},{"id":"https://openalex.org/C90673727","wikidata":"https://www.wikidata.org/wiki/Q901718","display_name":"Product (mathematics)","level":2,"score":0.3244999945163727},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.296099990606308},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.2946000099182129},{"id":"https://openalex.org/C148220186","wikidata":"https://www.wikidata.org/wiki/Q7111912","display_name":"Outcome (game theory)","level":2,"score":0.29409998655319214},{"id":"https://openalex.org/C33954974","wikidata":"https://www.wikidata.org/wiki/Q486494","display_name":"Sensor fusion","level":2,"score":0.2784000039100647},{"id":"https://openalex.org/C2776036281","wikidata":"https://www.wikidata.org/wiki/Q48769818","display_name":"Constraint (computer-aided design)","level":2,"score":0.26579999923706055},{"id":"https://openalex.org/C2776207758","wikidata":"https://www.wikidata.org/wiki/Q5303302","display_name":"Downstream (manufacturing)","level":2,"score":0.26429998874664307},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.2540000081062317},{"id":"https://openalex.org/C2779714256","wikidata":"https://www.wikidata.org/wiki/Q25305062","display_name":"Multiple Models","level":2,"score":0.2533000111579895}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.07779","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.07779","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":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.07779","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.07779","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":"article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Pathology":[0],"foundation":[1,28],"models":[2,29],"(FMs)":[3],"have":[4],"become":[5],"central":[6],"to":[7,117,175],"computational":[8],"histopathology,":[9],"offering":[10],"strong":[11],"transfer":[12],"performance":[13,161],"across":[14,99,162],"a":[15,31,59,105,181],"wide":[16],"range":[17],"of":[18,26],"diagnostic":[19],"and":[20,43,61,76,95,147,157],"prognostic":[21],"tasks.":[22],"The":[23,85],"rapid":[24],"proliferation":[25],"pathology":[27],"creates":[30],"model-selection":[32],"bottleneck:":[33],"no":[34,92,96],"single":[35,170],"model":[36,63],"is":[37,51,115],"uniformly":[38],"best,":[39],"yet":[40],"exhaustively":[41],"adapting":[42],"validating":[44],"many":[45],"candidates":[46],"for":[47],"each":[48],"downstream":[49],"endpoint":[50],"prohibitively":[52],"expensive.":[53],"We":[54,102,132],"address":[55],"this":[56],"challenge":[57],"with":[58,187],"lightweight":[60],"novel":[62],"fusion":[64,79,86,114],"strategy,":[65],"LogitProd,":[66],"which":[67],"treats":[68],"independently":[69],"trained":[70],"FM-based":[71,178],"predictors":[72],"as":[73,121,123],"fixed":[74],"experts":[75],"learns":[77],"sample-adaptive":[78],"weights":[80],"over":[81,167],"their":[82],"slide-level":[83],"outputs.":[84],"operates":[87],"purely":[88],"on":[89,136,154],"logits,":[90],"requiring":[91],"encoder":[93],"retraining":[94],"feature-space":[97],"alignment":[98],"heterogeneous":[100,177],"backbones.":[101],"further":[103],"provide":[104],"theoretical":[106],"analysis":[107],"showing":[108],"that":[109],"the":[110,124,129,159,168],"optimal":[111],"weighted":[112],"product":[113],"guaranteed":[116],"perform":[118],"at":[119],"least":[120],"well":[122],"best":[125],"individual":[126],"expert":[127],"under":[128],"training":[130,190],"objective.":[131],"systematically":[133],"evaluate":[134],"LogitProd":[135,151,172],"\\textbf{22}":[137],"benchmarks":[138],"spanning":[139],"WSI-level":[140],"classification,":[141,143],"tile-level":[142],"gene":[144],"mutation":[145],"prediction,":[146],"discrete-time":[148],"survival":[149],"modeling.":[150],"ranks":[152],"first":[153],"20/22":[155],"tasks":[156,164],"improves":[158],"average":[160],"all":[163],"by":[165],"~3%":[166],"strongest":[169],"expert.":[171],"enables":[173],"practitioners":[174],"upgrade":[176],"pipelines":[179],"in":[180],"plug-and-play":[182],"manner,":[183],"achieving":[184],"multi-expert":[185],"gains":[186],"$\\sim$12$\\times$":[188],"lower":[189],"cost":[191],"than":[192],"feature-fusion":[193],"alternatives.":[194]},"counts_by_year":[],"updated_date":"2026-04-14T06:02:45.956762","created_date":"2026-04-11T00:00:00"}