{"id":"https://openalex.org/W7225119587","doi":"https://doi.org/10.5281/zenodo.23895101","title":"Selective retrieval-augmented answering for medical multiple-choice QA with MedGemma-27B: results-only report (single-use public test) / RAG selettivo per QA medico a scelta multipla con MedGemma-27B: rapporto solo-risultati","display_name":"Selective retrieval-augmented answering for medical multiple-choice QA with MedGemma-27B: results-only report (single-use public test) / RAG selettivo per QA medico a scelta multipla con MedGemma-27B: rapporto solo-risultati","publication_year":2026,"publication_date":"2026-11-02","ids":{"openalex":"https://openalex.org/W7225119587","doi":"https://doi.org/10.5281/zenodo.23895101"},"language":"en","primary_location":{"id":"doi:10.5281/zenodo.23895101","doi":"https://doi.org/10.5281/zenodo.23895101","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.23895101","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.23895101","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5159643063","display_name":"Giovanni Del Papa","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Del Papa, Giovanni","raw_affiliation_strings":["Independent researcher, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Independent researcher, Italy","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5159643063"],"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/T10286","display_name":"Information Retrieval and Search Behavior","score":0.41190001368522644,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10286","display_name":"Information Retrieval and Search Behavior","score":0.41190001368522644,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10028","display_name":"Topic Modeling","score":0.32839998602867126,"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/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.15039999783039093,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/retrieval-augmented-generation","display_name":"retrieval-augmented generation","score":0.953000009059906},{"id":"https://openalex.org/keywords/confidence-gating","display_name":"confidence gating","score":0.7110000252723694},{"id":"https://openalex.org/keywords/passage-retrieval","display_name":"passage retrieval","score":0.47200000286102295},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.5509999990463257},{"id":"https://openalex.org/keywords/accuracy-improvement","display_name":"accuracy improvement","score":0.42399999499320984},{"id":"https://openalex.org/keywords/hallucination-mitigation","display_name":"hallucination mitigation","score":0.42100000381469727},{"id":"https://openalex.org/keywords/llm-as-a-judge","display_name":"LLM-as-a-judge","score":0.5239999890327454}],"concepts":[{"id":"https://openalex.org/C186282968","wikidata":"https://www.wikidata.org/wiki/Q1434261","display_name":"McNemar's test","level":2,"score":0.7001000046730042},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2989000082015991},{"id":"https://openalex.org/C2777267654","wikidata":"https://www.wikidata.org/wiki/Q3519023","display_name":"Test (biology)","level":2,"score":0.2888000011444092},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.28380000591278076},{"id":"https://openalex.org/C95457728","wikidata":"https://www.wikidata.org/wiki/Q309","display_name":"History","level":0,"score":0.2800000011920929},{"id":"https://openalex.org/C161191863","wikidata":"https://www.wikidata.org/wiki/Q199655","display_name":"Library science","level":1,"score":0.266400009393692},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.2662000060081482},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.26170000433921814},{"id":"https://openalex.org/C5274069","wikidata":"https://www.wikidata.org/wiki/Q2285707","display_name":"Categorical variable","level":2,"score":0.26030001044273376},{"id":"https://openalex.org/C113993141","wikidata":"https://www.wikidata.org/wiki/Q906759","display_name":"Marketing buzz","level":2,"score":0.25850000977516174},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2540000081062317},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.25130000710487366}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.5281/zenodo.23895101","doi":"https://doi.org/10.5281/zenodo.23895101","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.23895101","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.23895102","doi":"https://doi.org/10.5281/zenodo.23895102","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.23895102","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.23895101","doi":"https://doi.org/10.5281/zenodo.23895101","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.23895101","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":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":2,"referenced_works":["https://openalex.org/W7223659042","https://openalex.org/W7223700876"],"related_works":["https://openalex.org/W7204854250","https://openalex.org/W7156935348","https://openalex.org/W7155556934","https://openalex.org/W7164176609","https://openalex.org/W7167840872","https://openalex.org/W7134089510","https://openalex.org/W7165169326","https://openalex.org/W4402670290","https://openalex.org/W7223700876","https://openalex.org/W7172381002"],"abstract_inverted_index":{"English.":[0],"A":[1],"confidence-":[2],"and":[3,28,51,128,163,165,314],"relevance-gated":[4],"retrieval":[5,57],"layer":[6],"on":[7,21,48,67],"top":[8],"of":[9,72,168,173],"MedGemma-27B":[10,207],"(bf16,":[11],"vLLM,":[12],"greedy)":[13],"answers":[14],"with":[15,161],"retrieved":[16],"passages":[17,149],"only":[18,50,141],"for":[19,192],"questions":[20],"which":[22],"the":[23,121,134,156,166,169,174],"base":[24,122,278],"model":[25],"is":[26,129,137],"uncertain":[27],"a":[29,33,56,68,239,272],"Qwen3-4B":[30],"reranker":[31,219],"finds":[32],"highly":[34],"relevant":[35],"passage":[36],"(tc":[37],"=":[38,41,44,74,87,99,244,256,267],"0.8,":[39],"\u03c4":[40],"0.93245,":[42],"k":[43],"3;":[45],"thresholds":[46],"calibrated":[47],"train/validation":[49],"frozen":[52],"before":[53],"testing).":[54],"With":[55],"corpus":[58,136,227,286],"that":[59],"includes":[60,155],"non-commercially":[61],"licensed":[62,115],"sources":[63,116],"(\"full\"":[64],"variant),":[65],"accuracy":[66],"single-use":[69],"public":[70],"test":[71,237],"n":[73,243],"6,545":[75],"medical":[76],"multiple-choice":[77],"items":[78,102,107],"rises":[79],"from":[80],"70.15":[81],"%":[82,85,91,251,254,261],"to":[83,113],"71.00":[84],"(\u0394":[86,255],"+0.86":[88],"pp,":[89,124,258],"95":[90,260],"bootstrap":[92],"CI":[93,125],"[+0.55,":[94],"+1.16],":[95],"exact":[96],"McNemar":[97,264],"p":[98,266],"4.5\u00b710-8;":[100],"81":[101],"fixed,":[103],"25":[104],"broken;":[105],"245":[106],"gated).":[108],"The":[109,153],"same":[110],"pipeline":[111],"restricted":[112],"commercially":[114,139],"(\"comm\")":[117],"does":[118],"not":[119,138],"beat":[120],"(+0.08":[123],"[\u22120.41,":[126],"+0.55])":[127],"therefore":[130],"switched":[131],"off.":[132],"Because":[133],"effective":[135],"licensable,":[140],"these":[142],"aggregate":[143,310],"results":[144],"are":[145],"published:":[146],"no":[147],"index,":[148],"or":[150],"source":[151],"data.":[152],"report":[154],"formulas,":[157],"figures,":[158,307],"per-benchmark":[159],"tables":[160],"CIs":[162],"tests,":[164],"funnel":[167],"gating":[170],"stages.":[171],"Cost":[172],"run:":[175],"about":[176],"USD":[177],"0.32":[178],"(A100":[179],"80":[180],"GB,":[181],"0.227":[182],"h).":[183],"Hashes:":[184],"frozen.json":[185],"sha256":[186,189],"ee56a54b52b0fa178dca0f78d7665cfb158a7eeace8de171aec7100c5622ab99;":[187],"artifacts.tgz":[188],"8fb79b571d47fcc5f0c9bb8ed0225eb76153cdaa0bfa45bd9975180560f1e4b3.":[190],"Not":[191],"clinical":[193],"use.":[194],"Italiano.":[195],"Uno":[196],"strato":[197],"RAG":[198],"controllato":[199],"da":[200],"soglie":[201],"di":[202,242],"confidenza":[203],"e":[204],"rilevanza":[205],"sopra":[206],"risponde":[208],"con":[209],"i":[210],"documenti":[211],"solo":[212,293],"sulle":[213],"domande":[214],"incerte":[215],"per":[216],"cui":[217],"un":[218,221,226,236],"trova":[220],"passaggio":[222,299],"molto":[223],"rilevante.":[224],"Con":[225],"che":[228],"include":[229],"fonti":[230,273],"non":[231,275,288],"commerciali":[232,274],"(variante":[233],"\u00abcompleta\u00bb),":[234],"su":[235],"pubblico":[238],"uso":[240],"singolo":[241],"6.545":[245],"item":[246],"l\u2019accuratezza":[247],"passa":[248],"dal":[249],"70,15":[250],"al":[252],"71,00":[253],"+0,86":[257],"IC":[259],"[+0,55;":[262],"+1,16],":[263],"esatto":[265],"4,5\u00b710-8).":[268],"La":[269],"variante":[270],"limitata":[271],"supera":[276],"la":[277],"(+0,08":[279],"pp)":[280],"ed":[281],"\u00e8":[282,289],"spenta.":[283],"Poich\u00e9":[284],"il":[285],"efficace":[287],"commercializzabile,":[290],"si":[291],"pubblicano":[292],"questi":[294],"risultati":[295],"aggregati:":[296],"nessun":[297],"indice,":[298],"o":[300],"dato.":[301],"Files:":[302],"article_en.pdf,":[303],"article_it.pdf,":[304],"code.zip":[305],"(statistics,":[306],"LaTeX":[308],"sources,":[309],"results).":[311],"Licences:":[312],"article":[313],"figures":[315],"CC":[316],"BY":[317],"4.0;":[318],"code":[319],"Apache-2.0.":[320]},"counts_by_year":[],"updated_date":"2026-11-04T10:54:05.112534","created_date":"2026-11-03T00:00:00"}