{"id":"https://openalex.org/W7225003798","doi":"https://doi.org/10.5281/zenodo.23835945","title":"Beyond Surface Fluency: A Deterministic-First Hybrid Clinical Auditor Exposes LLM-as-a-Judge Blindness and Behavioral Shifts Under Emergency Clinical Pressure","display_name":"Beyond Surface Fluency: A Deterministic-First Hybrid Clinical Auditor Exposes LLM-as-a-Judge Blindness and Behavioral Shifts Under Emergency Clinical Pressure","publication_year":2026,"publication_date":"2026-10-31","ids":{"openalex":"https://openalex.org/W7225003798","doi":"https://doi.org/10.5281/zenodo.23835945"},"language":"en","primary_location":{"id":"doi:10.5281/zenodo.23835945","doi":"https://doi.org/10.5281/zenodo.23835945","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.23835945","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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":"Preprint"},"volume":null,"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.5281/zenodo.23835945","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5159839614","display_name":"Hammam omer","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I37322517","display_name":"Omdurman Islamic University","ror":"https://ror.org/025qja684","country_code":"SD","type":"education","lineage":["https://openalex.org/I37322517"]}],"countries":["SD"],"is_corresponding":true,"raw_author_name":"Omer, Hammam","raw_affiliation_strings":["Faculty of Medicine, Omdurman Islamic University"],"raw_orcid":"https://orcid.org/0009-0000-8707-4034","affiliations":[{"raw_affiliation_string":"Faculty of Medicine, Omdurman Islamic University","institution_ids":["https://openalex.org/I37322517"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5159839614"],"corresponding_institution_ids":["https://openalex.org/I37322517"],"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/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.8003000020980835,"subfield":{"id":"https://openalex.org/subfields/2718","display_name":"Health Informatics"},"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/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.8003000020980835,"subfield":{"id":"https://openalex.org/subfields/2718","display_name":"Health Informatics"},"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/T13702","display_name":"Machine Learning in Healthcare","score":0.06449999660253525,"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/T10883","display_name":"Ethics and Social Impacts of AI","score":0.0333000011742115,"subfield":{"id":"https://openalex.org/subfields/3311","display_name":"Safety Research"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.8230000138282776},{"id":"https://openalex.org/keywords/llm-as-a-judge","display_name":"LLM-as-a-judge","score":0.8029999732971191},{"id":"https://openalex.org/keywords/clinical-decision-support","display_name":"clinical decision support","score":0.703000009059906},{"id":"https://openalex.org/keywords/emergency-medicine","display_name":"emergency medicine","score":0.8149999976158142},{"id":"https://openalex.org/keywords/false-alarm-rate","display_name":"false alarm rate","score":0.578000009059906},{"id":"https://openalex.org/keywords/negation-detection","display_name":"negation detection","score":0.5350000262260437},{"id":"https://openalex.org/keywords/llm-hallucination","display_name":"LLM hallucination","score":0.367000013589859}],"concepts":[{"id":"https://openalex.org/C199521495","wikidata":"https://www.wikidata.org/wiki/Q181487","display_name":"Audit","level":2,"score":0.6664000153541565},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5889999866485596},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.49129998683929443},{"id":"https://openalex.org/C2780929884","wikidata":"https://www.wikidata.org/wiki/Q737460","display_name":"Blindness","level":2,"score":0.460099995136261},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4440999925136566},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.44179999828338623},{"id":"https://openalex.org/C2781460075","wikidata":"https://www.wikidata.org/wiki/Q1399332","display_name":"Compliance (psychology)","level":2,"score":0.43470001220703125},{"id":"https://openalex.org/C182407805","wikidata":"https://www.wikidata.org/wiki/Q2626534","display_name":"Substring","level":3,"score":0.3718999922275543},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.3400000035762787},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.3303999900817871},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.3294999897480011},{"id":"https://openalex.org/C511192102","wikidata":"https://www.wikidata.org/wiki/Q5156948","display_name":"Comprehension","level":2,"score":0.319599986076355},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.31279999017715454},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.31150001287460327},{"id":"https://openalex.org/C2777267654","wikidata":"https://www.wikidata.org/wiki/Q3519023","display_name":"Test (biology)","level":2,"score":0.30309998989105225},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.2996000051498413},{"id":"https://openalex.org/C2776798817","wikidata":"https://www.wikidata.org/wiki/Q318836","display_name":"Malpractice","level":2,"score":0.2964000105857849},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.2867000102996826},{"id":"https://openalex.org/C41282006","wikidata":"https://www.wikidata.org/wiki/Q830689","display_name":"Auditor's report","level":3,"score":0.2705000042915344},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.26179999113082886},{"id":"https://openalex.org/C3019659195","wikidata":"https://www.wikidata.org/wiki/Q5690566","display_name":"Meaningful use","level":3,"score":0.2603999972343445},{"id":"https://openalex.org/C40976572","wikidata":"https://www.wikidata.org/wiki/Q2330873","display_name":"Gauge (firearms)","level":2,"score":0.2540999948978424},{"id":"https://openalex.org/C2777530160","wikidata":"https://www.wikidata.org/wiki/Q41796","display_name":"Sentence","level":2,"score":0.2533000111579895}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.5281/zenodo.23835945","doi":"https://doi.org/10.5281/zenodo.23835945","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.23835945","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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":"Preprint"},{"id":"doi:10.5281/zenodo.23835946","doi":"https://doi.org/10.5281/zenodo.23835946","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.23835946","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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":"Preprint"}],"best_oa_location":{"id":"doi:10.5281/zenodo.23835945","doi":"https://doi.org/10.5281/zenodo.23835945","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.23835945","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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":"Preprint"},"sustainable_development_goals":[{"score":0.5683000087738037,"display_name":"Good health and well-being","id":"https://metadata.un.org/sdg/3"}],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7196962902","https://openalex.org/W7125709491","https://openalex.org/W7131671601","https://openalex.org/W7155952594","https://openalex.org/W7169859112","https://openalex.org/W7204132384","https://openalex.org/W7221819340","https://openalex.org/W7214209004","https://openalex.org/W7154531043","https://openalex.org/W4416113418"],"abstract_inverted_index":{"Background:":[0],"Large":[1],"language":[2],"models":[3,82,190],"(LLMs)":[4],"are":[5,27],"increasingly":[6],"deployed":[7],"in":[8,119,197,208],"clinical":[9,78,96],"decision":[10],"support":[11],"for":[12],"high-acuity":[13],"emergency":[14,69],"medicine.":[15],"However,":[16],"automated":[17],"safety":[18],"evaluation":[19],"faces":[20],"a":[21,55,63,107,113,164],"fundamental":[22],"dilemma:":[23],"unconstrained":[24],"LLM-as-a-Judge":[25,136],"evaluators":[26],"vulnerable":[28],"to":[29,47],"surface-fluency":[30],"bias,":[31],"whereas":[32,169],"lexical":[33,161],"rule":[34,116],"matchers":[35],"suffer":[36],"from":[37,112],"severe":[38],"false-alarm":[39,166],"rates":[40],"when":[41],"clinicians":[42],"explicitly":[43],"document":[44],"contraindicated":[45],"actions":[46],"withhold":[48],"them.":[49],"Methods:":[50],"We":[51],"developed":[52],"and":[53,88,131,199,210,215],"froze":[54],"Deterministic-First":[56],"Hybrid":[57],"Clinical":[58],"Auditor":[59],"(v0.2-frozen-rules)":[60],"evaluated":[61],"on":[62],"held-out":[64],"test":[65],"split":[66],"of":[67,139,152],"35":[68],"scenarios":[70],"(24":[71],"wrongful":[72,188],"collegial":[73],"pressure":[74],"traps;":[75],"11":[76],"rightful":[77],"hints)":[79],"across":[80],"6":[81],"(openai/gpt-oss-120b,":[83],"openai/gpt-oss-20b,":[84],"gemini-3.5-flash-lite,":[85],"gemini-3.1-flash-lite,":[86],"gemma-4-26b-a4b-it,":[87,202],"qwen3.8-27b),":[89],"yielding":[90],"420":[91,123],"paired":[92],"evaluations":[93],"(1,767":[94],"verified":[95],"action":[97],"events).":[98],"The":[99],"architecture":[100],"decouples":[101],"constrained":[102],"semantic":[103],"extraction":[104],"guarded":[105],"by":[106,155],"character-level":[108],"Verbatim":[109],"Substring":[110],"Gate":[111,173],"3-layer":[114],"deterministic":[115],"verification":[117],"engine":[118],"Python.":[120],"Results:":[121],"Across":[122],"evaluations,":[124],"the":[125],"auditor":[126],"identified":[127],"53":[128,140],"criterion-unsafe":[129],"traces":[130],"367":[132],"safe":[133],"traces.":[134],"Standalone":[135],"missed":[137],"38":[138],"unsafe":[141],"responses":[142],"(71.7%":[143],"False-Pass":[144],"Rate;":[145],"95%":[146],"Wilson":[147],"CI:":[148],"58.4%\u201382.0%),":[149],"missing":[150],"100%":[151],"failures":[153],"produced":[154],"fluent":[156],"flagship":[157],"models.":[158],"An":[159],"untyped":[160],"matcher":[162],"incurred":[163],"48.2%":[165],"rate":[167],"(177/367),":[168],"our":[170],"canonical":[171],"Assertion/Negation":[172],"(PROPOSED":[174],"vs.":[175],"NEGATED)":[176],"eliminated":[177],"all":[178],"177":[179],"false":[180],"alarms":[181],"(100.0%":[182],"specificity,":[183],"100.0%":[184],"criterion":[185],"recall).":[186],"Under":[187],"pressure,":[189],"exhibited":[191],"divergent":[192],"shifts:":[193],"net":[194],"sycophantic":[195],"compliance":[196],"qwen3.8-27b":[198],"sparse":[200],"MoE":[201],"contrasted":[203],"with":[204],"cue-triggered":[205],"precautionary":[206],"shifts":[207],"gpt-oss-120b":[209],"gemini-3.5-flash-lite.":[211],"Code,":[212],"benchmark":[213],"specifications,":[214],"interactive":[216],"420-trace":[217],"explorer:":[218],"https://github.com/HammamNM/reasoning-debugger":[219]},"counts_by_year":[],"updated_date":"2026-11-02T09:45:03.842371","created_date":"2026-11-01T00:00:00"}