{"id":"https://openalex.org/W7153126676","doi":"https://doi.org/10.48550/arxiv.2604.07936","title":"Shortcut Learning in Glomerular AI: Adversarial Penalties Hurt, Entropy Helps","display_name":"Shortcut Learning in Glomerular AI: Adversarial Penalties Hurt, Entropy Helps","publication_year":2026,"publication_date":"2026-04-09","ids":{"openalex":"https://openalex.org/W7153126676","doi":"https://doi.org/10.48550/arxiv.2604.07936"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.07936","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.07936","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.2604.07936","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5029561767","display_name":"Mohammad Daouk","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Daouk, Mohammad","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125755577","display_name":"Jan Ulrich Becker","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Becker, Jan Ulrich","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025494644","display_name":"Neeraja Kambham","orcid":"https://orcid.org/0000-0003-1136-3686"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kambham, Neeraja","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133367516","display_name":"Anthony Chang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chang, Anthony","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133386864","display_name":"Hien Nguyen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Van Nguyen, Hien","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5133345317","display_name":"Chandra Mohan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mohan, Chandra","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5029561767"],"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/T11438","display_name":"Retinal Imaging and Analysis","score":0.273499995470047,"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/T11438","display_name":"Retinal Imaging and Analysis","score":0.273499995470047,"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/T10862","display_name":"AI in cancer detection","score":0.1429000049829483,"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/T11289","display_name":"Single-cell and spatial transcriptomics","score":0.04960000142455101,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/stain","display_name":"Stain","score":0.8791000247001648},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.4927000105381012},{"id":"https://openalex.org/keywords/entropy","display_name":"Entropy (arrow of time)","score":0.40310001373291016},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.3986999988555908},{"id":"https://openalex.org/keywords/lesion","display_name":"Lesion","score":0.36059999465942383},{"id":"https://openalex.org/keywords/bayesian-probability","display_name":"Bayesian probability","score":0.32600000500679016},{"id":"https://openalex.org/keywords/grading","display_name":"Grading (engineering)","score":0.29829999804496765}],"concepts":[{"id":"https://openalex.org/C2781294515","wikidata":"https://www.wikidata.org/wiki/Q2733470","display_name":"Stain","level":3,"score":0.8791000247001648},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5813999772071838},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5461000204086304},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.4927000105381012},{"id":"https://openalex.org/C106301342","wikidata":"https://www.wikidata.org/wiki/Q4117933","display_name":"Entropy (arrow of time)","level":2,"score":0.40310001373291016},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.3986999988555908},{"id":"https://openalex.org/C2781156865","wikidata":"https://www.wikidata.org/wiki/Q827023","display_name":"Lesion","level":2,"score":0.36059999465942383},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.32600000500679016},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.30480000376701355},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3043999969959259},{"id":"https://openalex.org/C2777286243","wikidata":"https://www.wikidata.org/wiki/Q5591926","display_name":"Grading (engineering)","level":2,"score":0.29829999804496765},{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.27720001339912415},{"id":"https://openalex.org/C19499675","wikidata":"https://www.wikidata.org/wiki/Q232207","display_name":"Monte Carlo method","level":2,"score":0.27480000257492065},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.26499998569488525},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.258899986743927},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.25870001316070557},{"id":"https://openalex.org/C167981619","wikidata":"https://www.wikidata.org/wiki/Q1685498","display_name":"Cross entropy","level":3,"score":0.25699999928474426},{"id":"https://openalex.org/C160234255","wikidata":"https://www.wikidata.org/wiki/Q812535","display_name":"Bayesian inference","level":3,"score":0.2567000091075897},{"id":"https://openalex.org/C176641082","wikidata":"https://www.wikidata.org/wiki/Q2446767","display_name":"Spectral signature","level":2,"score":0.2556000053882599}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.07936","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.07936","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.2604.07936","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.07936","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":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Stain":[0],"variability":[1],"is":[2,120],"a":[3,28,43,88,102,124,183,196,205],"pervasive":[4],"source":[5],"of":[6,47,135],"distribution":[7],"shift":[8],"and":[9,30,58,74,94,100,133,195],"potential":[10,210],"shortcut":[11,152],"learning":[12,153],"in":[13,81,213],"renal":[14],"pathology":[15],"AI.":[16,215],"We":[17,41,70],"ask":[18],"whether":[19],"lupus":[20],"nephritis":[21],"glomerular":[22,49,214],"lesion":[23,93,144,178],"classifiers":[24],"exploit":[25],"stain":[26,37,95,98,107,114,118,136,140,162,172,193],"as":[27,66],"shortcut,":[29],"how":[31],"to":[32,192],"mitigate":[33],"such":[34],"bias":[35],"without":[36,176],"or":[38,180],"site":[39],"labels.":[40],"curate":[42],"multi-center,":[44],"multi-stain":[45,186],"dataset":[46,187],"9,674":[48],"patches":[50],"(224$\\times$224)":[51],"from":[52],"365":[53],"WSIs":[54],"across":[55],"three":[56,82],"centers":[57],"four":[59],"stains":[60],"(PAS,":[61],"H&E,":[62],"Jones,":[63],"Trichrome),":[64],"labeled":[65],"proliferative":[67],"vs.":[68],"non-proliferative.":[69],"evaluate":[71],"Bayesian":[72,197],"CNN":[73],"ViT":[75],"backbones":[76],"with":[77,96,105,200],"Monte":[78],"Carlo":[79],"dropout":[80],"settings:":[83],"(1)":[84],"stain-only":[85],"classification;":[86],"(2)":[87],"dual-head":[89,103,198],"model":[90,104],"jointly":[91],"predicting":[92],"supervised":[97],"loss;":[99],"(3)":[101],"label-free":[106,201],"regularization":[108,170,203],"via":[109],"entropy":[110,202],"maximization":[111],"on":[112,154],"the":[113,131],"head.":[115],"In":[116,128,167],"(1),":[117],"identity":[119],"trivially":[121],"learnable,":[122],"confirming":[123],"strong":[125],"candidate":[126],"shortcut.":[127],"(2),":[129],"varying":[130],"strength":[132],"sign":[134],"supervision":[137],"strongly":[138],"modulates":[139],"performance":[141],"but":[142],"leaves":[143],"metrics":[145],"essentially":[146],"unchanged,":[147],"indicating":[148],"no":[149],"measurable":[150],"stain-driven":[151],"this":[155],"multi-stain,":[156],"multi-center":[157],"dataset,":[158],"while":[159],"overly":[160],"adversarial":[161],"penalties":[163],"inflate":[164],"predictive":[165],"uncertainty.":[166],"(3),":[168],"entropy-based":[169],"holds":[171],"predictions":[173],"near":[174],"chance":[175],"degrading":[177],"accuracy":[179],"calibration.":[181],"Overall,":[182],"carefully":[184],"curated":[185],"can":[188],"be":[189],"inherently":[190],"robust":[191],"shortcuts,":[194],"architecture":[199],"offers":[204],"simple,":[206],"deployment-friendly":[207],"safeguard":[208],"against":[209],"stain-related":[211],"drift":[212]},"counts_by_year":[],"updated_date":"2026-04-14T06:02:45.956762","created_date":"2026-04-11T00:00:00"}